The Politics of Incommensurability: The Case of Rorty and Foucault
By
Paul Showler
A thesis submitted in partial fulfillment of the requirements for the degree of
Master of Arts
Department of Philosophy
University of Alberta
© Paul Showler, 2016
ii
Abstract
Despite their extensive philosophical agreements, the case can be made that Richard
Rorty and Michel Foucault were at odds politically. The devout liberal that he was, Rorty
struggled continuously to come to grips with what he took to be a sort of radicalism espoused by
the French thinker. As a result, Rorty’s critical engagement with Foucault over political matters
was always two-sided. On the one hand, he commended the latter’s ability to draw attention to
the insidious aspects of contemporary liberal societies. On the other hand, he was critical of
Foucault’s apparent disdain for liberal societies and institutions. Though Rorty would often
frame this criticism in different ways, he resisted viewing his political disagreement with
Foucault as having any philosophical basis. That is, he tended to see Foucault’s radicalism as the
result of contingent sociological and biographical factors, rather than as the outcome of
philosophical argument. Against this picture, the aim of my thesis is to explain the political
disagreement between Foucault and Rorty on the basis of their divergent views concerning
conceptual relativism—the idea that there can exist incommensurable conceptual frameworks. In
particular, I suggest that whereas Rorty’s pragmatic defense of liberalism is bolstered by his
rejection of conceptual relativism as a tenable philosophical thesis, many of Foucault’s more
radical-sounding political claims are supported by the version of conceptual relativism that he
put forth in his archaeological writings of the 1960’s.
iii
I dedicate this work to my parents.
iv
Acknowledgments
I am very grateful to Chloë Taylor for her unwavering encouragement and countless
comments throughout every stage of this project. Her mentorship has played an enormous role in
my intellectual and professional development. I am also grateful to Bruce Hunter and Ingo
Brigandt for providing such stimulating and insightful comments on earlier drafts. It is
impossible for me to imagine a better supervisory committee. I owe a great deal to everyone in
the Department of Philosophy at the University of Alberta for their friendship and support.
I would not have been able to undertake this project without the financial assistance of a
SSHRC Joseph-Armand Bombardier Canada Graduate Scholarship, Walter H. Johns Graduate
Fellowship, Queen Elizabeth II Graduate Scholarship, as well as funding from the Department of
Philosophy at the University of Alberta.
v
Abbreviations
The following abbreviations are used in citing works by Michel Foucault.
LCP: Language, Counter-Memory, Practice: Selected Essays and Interviews
PK: Power/Knowledge
STP: Security, Territory, Population: Lectures at the Collège de France 1977-8
AK: The Archaeology of Knowledge
BB: The Birth of Biopolitics: Lectures at the Collège de France 1978-9
FR: The Foucault Reader
HS: The History of Sexuality Volume 1: An Introduction
OT: The Order of Things: An Archaeology of the Human Sciences
The following abbreviations are used in citing works by Richard Rorty.
AOC: Achieving Our Country
CP: Consequences of Pragmatism
CIS: Contingency, Irony, and Solidarity
EHO: Essays on Heidegger and Others
ORT: Objectivity, Relativism and Truth
PMN: Philosophy and the Mirror of Nature
TP: Truth and Progress
vi
Contents
Introduction 1
1 Rorty and Foucault’s Political Disagreement 8
1.1 Postmodern Bourgeois Liberalism and the Public-Private Split 9
1.2 Rorty’s Disagreement with Foucault 19
2 Conceptual Relativism 28
2.1 Foucault and Epistemology 30
2.2 Rorty and the Davidsonian Attack on Conceptual Relativism 37
2.3 Incommensurability in Foucault 44
3 Politics of Incommensurability 56
3.1 Liberal Progress: Incommensurable Views 56
3.2 On the Very Idea of Sketching Utopias 68
Conclusion 79
Bibliography 86
1
Introduction
Richard Rorty and Michel Foucault occupy a lot of philosophical common ground.
Neither saw much of a point in doing metaphysics or epistemology, as conceived by the Western
philosophical tradition; both thinkers shared Nietzsche’s suspicion of traditional attempts to
uncover a common human essence or nature—especially insofar as such attempts are made in the
name of guiding or justifying socio-political projects; and a sensitivity to historical contingency
runs through both of their thought—they agree that truth, knowledge, and even our self-
conceptions are bound up with historically evolving linguistic practices and social institutions.
Politically, however, Rorty’s ‘postmodern’ brand of reformist liberalism seems to stand
in stark contrast with a discernable radicalism in many of Foucault’s writings—especially in
interviews from the early seventies. Indeed, because of this apparent political disagreement,
Rorty’s critical engagement with the French thinker was almost always two-sided. On the one
hand, he viewed Foucault not only as a fellow pragmatist, helping to overturn the metaphysico-
epistemological picture that has been holding philosophers captive for so long; but, more
importantly, he saw him as a politically minded intellectual who did as much as anyone to draw
attention to the insidious aspects of contemporary societies. On the other hand, Rorty was critical
of what he took to be Foucault’s unwillingness to be a liberal. Indeed, there are several places
(mostly in interviews) where Foucault seems to suggest that liberal societies and their institutions
are too problematic to be reformed; that they must be replaced with something not only entirely
different, but from our current standpoint, unthinkable. Such remarks clearly perturbed Rorty. So
much so, that his many discussions of Foucault seem to be driven by an almost pathological urge
to explain the latter’s radicalism, more so than refute it. As Wojeck Malecki (2011) has recently
pointed out, Rorty would frequently adopt different interpretive strategies, intended to make
2
Foucault’s ‘radicalism’ seem like the product of contingent sociological factors, or the result of
idiosyncratic experiences with historical archives and leather bars. In other words, rather than
attributing philosophical reasons as the source of their political disagreement over liberalism,
Rorty tended to see it as having some ‘external’ source. This approach is put quite bluntly in an
interview where he writes,
I don’t see what Foucault had against bourgeois liberalism, except in France of the fifties
and sixties it just wasn’t respectable to be a bourgeois liberal. I don’t think he has any
arguments against it or anything better to suggest. So I’m inclined to think that his
opposition to liberalism and reformism was merely a contingent French fashion (TCF,
30).
The aim of my thesis is to suggest a novel approach for framing the political
disagreement between Foucault and Rorty. Without downplaying either the role that dissimilar
socio-political milieus played in shaping their political views, or the extensive philosophical
common ground between them, I shall identify what I take to be a significant, but overlooked
philosophical issue on which they are at odds. I shall then try to show how this philosophical
disagreement can help explain their political differences.
Philosophically, I take Rorty and Foucault to be at odds over a view called conceptual
relativism—which is the view that it is possible for two or more (for lack of a better term)
‘discursive frameworks’ to be incommensurable. Thanks to a series of arguments put forth by the
American philosopher Donald Davidson, Rorty is convinced that we should reject the idea that
two languages, conceptual schemes, or ‘discursive frameworks’, can be incommensurable. This
is especially true if by incommensurable, we mean ‘untranslatable.’ Although, as far as I know,
Foucault never uses the term ‘incommensurable’, I shall suggest that he is committed to a form
of conceptual relativism, which is most apparent in his archaeological writings of the 1960s. My
central claim is thus two-fold: while Rorty’s skepticism towards conceptual relativism can help
3
explain why liberalism seemed so plausible (if not inevitable) to him, Foucault’s willingness to
embrace the doctrine can help explain both his reluctance to praise liberal societies in
comparison to past ones, as well as his refusal to offer positive political proposals about how
they might be improved (or even replaced). Before presenting a roadmap of my arguments, I
shall explain why I think this is a valuable project, as well as alert the reader to some of its
limitations.
In general, only a handful of commentators have taken a serious look at the political
intersection between the Rorty and Foucault.1 While this can partially be explained by the fact
that many of Rorty’s rhetorical strategies for framing his criticisms of Foucault can easily be
dismissed as misreadings, along with the fact that Foucault only mentions Rorty once in the
entirety of his publications, I find this paucity of dialogue surprising. Surprising, because unlike
almost all of Foucault’s other ‘liberal critics’ (to borrow a term from David Halperin), Rorty’s
metaphilosophical views are often virtually the same as Foucault’s. So often, arguments directed
at Foucault from the liberal standpoint proceed by insisting that his anti-essentialism precludes
the possibility of projects of human emancipation.2 Precisely because Rorty has no recourse to
such a strategy, his political disagreement with Foucault is unique—it proceeds on
metaphilosophical terms with which Foucault would be more inclined to agree.
In particular, I hope that framing this political disagreement in terms of the question of
conceptual relativism will prove fruitful. While on at least one occasion, Rorty explicitly links
his defense of liberalism to the impossibility of conceptual relativity, as far as I am aware, no
other commentator has attempted to pursue this connection further. Should I succeed in showing
1 Allen (2009, 2010); Malecki (2011); Koopman (2013).
2 I think that Ian Hacking manages to cut to the core of this common strategy by imagining a conversation between
Foucault and Kant: “What is man?” asked Kant. Nothing, says Foucault. “For what then may we hope?” asked Kant.
Does Foucault give the same nothing in reply? (HO, 86).
4
that Rorty’s pragmatic defense of liberalism depends on his commitment to the view that
conceptual relativism is false, this would be an important advancement in understanding his
thought. Likewise, while Foucault’s commitment to conceptual relativism has been discussed, I
am unaware of any other attempts to establish a connection between it and his political views. 3
Finally, while I hope that drawing attention to this overlooked philosophical difference
between Foucault and Rorty as a means of explaining their political disagreement will open up a
space for dialogue between contemporary scholars, I am also interested in the possibility of
generalizing this insight. For instance, can one use the notion of conceptual relativism in order to
articulate the difference between radicalism and reformism? Might we say that the rejection of
the very possibility of incommensurability (in one of its many senses), is a condition for having a
reformist politics, or attitude? Conversely, can one be said to be radical without recognizing
some form of incommensurability between the social, political, ethical, or epistemological
framework that one seeks to transform, and that which one expects to arise as a result of the
transformation? One reason that it might be especially useful to theorize the difference between
reformism and radicalism is that the notion of radicalism has become extremely problematized in
Western societies. North American and European governments are increasingly concerned with
the so-called ‘radicalization of youth,’ or the ‘radicalization of religious sects.’ There are obvious
dangers in allowing these notions to influence or dictate social policy without attempting to
clarify what, precisely, is at stake when we speak of radicalization.4
3 See ‘Michel Foucault’s Immature Science’ by Ian Hacking, In Historical Ontology (2002).
4 A more robust account of the connection between conceptual relativism and political attitudes could also include a
discussion of conservativism. Interestingly, Foucault has been labeled as a ‘young conservative’ by Habermas
(1981). Though Habermas’s charge centers on tensions in Foucault’s purported critique of modernity, one might
develop another line of thought which takes the position expressed in Foucault’s archaeological writings to leave
him unable to account for the ways in which agents can effect radical change or even reform. Even if, however,
such an interpretation holds, still more work would need to be done in order to show that this inability gives rise to
the normative claim that certain institutions, and systems of values ought to be preserved. My motivation for
5
Regarding the scope of my project, there are some caveats that should be mentioned at
the outset. First, my account of Foucault is heavily weighted toward his archaeological period; in
particular, it draws most frequently from The Order of Things and The Archaeology of
Knowledge. There are several reasons for this decision. First, it is in these writings that
Foucault’s commitment to conceptual relativism is most apparent. Second, the characterization
of Foucault as adopting a radical attitude towards politics is most strongly supported by remarks
he made in the early 1970’s. Indeed, the key passages which Rorty initially cites in support of his
political disagreement with Foucault occur in interviews from 1971. Since these remarks are in
such close proximity to Foucault’s archaeological writings, and my aim is to develop a
connection between the two, it only seems appropriate to focus mainly on these writings. This
decision, of course, raises many questions. For instance, did Foucault continue to adopt a radical
attitude towards politics throughout his entire career, or was it only a temporary phenomenon?
Moreover, was Foucault’s commitment to conceptual relativism only a function of his
archaeological writings, or did he abandon it in his later work? Answering these questions would
require an extremely broad account of Foucault’s intellectual trajectory, which is unfortunately
beyond the scope of this project. It would, for instance, require an investigation into the
relationship that Foucault’s archaeological writings have to his later genealogical and ethical
projects, as well as a comparative study of his political commitments. Fortunately, these
questions have been pursued extensively in the literature; unfortunately, there is little consensus
about how to answer them. Therefore, my claim that the political disagreement between Rorty
and Foucault can be explained on the basis of their disagreement over the plausibility of
conceptual relativism should be understood with these qualifications in mind.
limiting the discussion to reformism and radicalism is that this is how Rorty typically frames his disagreement with
Foucault.
6
Second, I shall plead agnosticism regarding both the plausibility of Rorty and Foucault’s
views, as well as over the question of their political disagreement. I consider both thinkers to be
among the most original and important of the 20th
century, but I also recognize that they are
among the most controversial. My aim is not to defend either of their views against any of their
myriad critics, but simply to present, as charitably as possible, what I take to be their political
disagreement, and offer, what I take to be a plausible yet overlooked explanation of it.
My thesis is divided into three chapters. In Chapter One I present an overview of Rorty’s
political position, as well as my own account of the political criticism of Foucault to which it
gives rise. On my reading, Rorty’s complaint is two-fold: on one hand, Foucault is unwilling to
commend liberal societies in comparison to those of the past (i.e., admit of progress), on the
other, he is unwilling to offer positive proposals for how to improve or replace current societies
(i.e., sketch utopias). Finally, following a recent study by Wojciech Malecki, I suggest that Rorty
did not take his political disagreement with Foucault to be grounded in philosophical reasons.
In Chapter Two, I argue that Foucault and Rorty are at odds over the question of
conceptual relativism. After explaining how Rorty came to reject the very idea of
incommensurable conceptual schemes through the work of Donald Davidson, I develop the
claim that Foucault was in fact committed to such a view. Through a close reading of the first
half of The Order of Things, I contend that Foucault’s commitment to conceptual relativism
results from (or amounts to) his claim that epistemes—roughly, an underlying system of rules
that define the historical conditions of possibility for positive bodies of knowledge—undergo
ruptures, or sharp discontinuities, which make them incommensurable with one another.
In Chapter Three, I argue that this disagreement over the question of conceptual
relativism can help explain the two men’s political disagreement. My strategy will be to examine
7
independently, both ‘sides’ of Rorty’s twofold criticism. I shall argue that Rorty’s pragmatic
defense of liberalism depends on our ability to place such societies and institutions within
concrete comparisons with other social forms (past, present and future), and that this strategy is
implausible unless we can, in advance, rule out the possibility that such comparisons might be
incommensurable (and thus, in some sense, invalid). By contrast, I shall suggest that Foucault’s
reluctance to praise liberal societies and institutions as instances of progress stems from his
conceptual relativism. My argument is that his openness to the incommensurability between
epistemes led Foucault to reject the idea that one could characterize as instances of progress, the
movement from body of knowledge to another (when either body is dependent on a disparate
episteme). I shall suggest that since Foucault allowed for the expansion of his archaeological
method to political domains, we can infer by analogy that similar considerations likely motivated
his reluctance to treat changes in social forms as questions of progress. To make this claim more
plausible, I develop some concrete examples from Foucault’s writings on ‘governmenality’ from
the late 1970s. Finally, I shall address the question of Foucault’s reluctance, as Rorty puts it, ‘to
sketch utopias’. While I do think that a large part of this disagreement is merely apparent—and
ultimately depends on how some of Foucault’s remarks are interpreted—I shall argue that if one
works out the full implications of Foucault’s conceptual relativism, certain problems emerge that
pose dangers for both radical and reformist attempts to sketch positive political programs.
8
Chapter One: Rorty and Foucault’s Political Disagreement: On What Grounds?
Despite their extensive philosophical agreements, the case can be made that Rorty and
Foucault are at odds politically. While the extent of this disagreement would be virtually
impossible to establish definitively—there are certain passages in Foucault’s writings as well as
in interviews which lend themselves to a more radical interpretation of his political views than
others—it is clear that there is a perceived difference between them, at least from Rorty’s
perspective. The devout liberal that he was, Rorty struggled continuously to come to grips with
what he took to be a radical streak in the French thinker. As a result, Rorty’s critical engagement
with Foucault over political matters is always two-sided. On the one hand, he commends
Foucault’s ability to draw attention to the insidious aspects of contemporary liberal societies. On
the other hand, he was critical of (1) Foucault’s reluctance to praise liberal societies for their
benefits, as well as (2) his unwillingness to offer proposals about what a better society might
look like.5 While Rorty’s delivery of these criticisms is often couched in idiosyncrasies, I do not
consider them to be particularly unique complaints about Foucault.
I begin this chapter by sketching what can be thought of as Rorty’s metaphilosophical
political stance—what he calls ‘postmodern bourgeois liberalism’—the view that we should
work to preserve the institutions of liberal democratic societies, without attempting to justify
them on the basis of ahistorical, universalist premises. Since Rorty frames his political
disagreement with Foucault with reference to this metaphilosophical stance, it shall serve as the
context in which I will initially present their disagreement. After that I shall present a survey of
5 For example Rorty writes, ‘A large part of Foucault’s work—the most valuable part, in my view—consists in
showing how the patterns of acculturation characteristic of liberal societies have imposed on their members kinds of
constraints of which older, premodern societies had not dreamed. He is not, however, willing to see these constraints
as compensated for by a decrease in pain, any more than Nietzsche was willing to see the resentfulness of “slave-
morality” as compensated for by such a decrease’ (CIS, 63).
9
some recent commentaries devoted to Rorty’s engagement with Foucault in particular, as well as
the connection between the former’s philosophical views and his liberalism.
Ultimately, the purpose of this chapter is to establish what might be thought of as the
received view regarding Rorty’s critique of Foucault. On this view, there is no substantial
philosophical disagreement between the two thinkers on the basis of which to attribute their
political divergence (i.e., Rorty’s reformist liberalism, and Foucault’s radicalism). Not only do I
take this to be the view presupposed by recent commentators, but I also believe that it was what
Rorty thought. In subsequent chapters I shall argue that this view overlooks an important
philosophical disagreement between the two thinkers (over the question of incommensurability),
and that this disagreement can, in turn, offer an alternative explanation as to why they were at
odds politically.
1.1 Postmodern Bourgeois Liberalism and the Public-Private Split
Around the middle of the 1980’s political questions began to emerge at the forefront of
Richard Rorty’s writings. The final section of his collection of essays Objectivity, Relativism,
and Truth deals not so much with his critiques of representationalism, epistemological
foundationalism, and metaphysics, but attempts to sketch what might be thought of as a
metaphilosophical view about politics—a view he calls postmodern bourgeois liberalism.6
Despite his eventual repudiation of the term7, Rorty, for whatever reason, chose to initially
describe his own views as postmodern. According to Barry Allen, the ‘simplest, most helpful
6 Years later, when asked about the title in an interview, Rorty responded by saying that ‘[t]hat designation
[“postmodern bourgeois liberal”] was supposed to be a joke. I thought it was a cute oxymoron—but no one else
seemed to think it was funny’ (TCF, 52). 7 Cf. Rorty’s remarks in ‘Pragmatism and post-Nietzschean Philosophy’ where he writes: ‘I have sometimes used
“postmodern” myself, in the rather narrow sense defined by Lyotard… But I now wish I had not. The term has been
so over-used that it is causing more trouble than it is worth’ (1).
10
meaning for the expression ‘postmodern’ refers to trends in German and French thought, and
now international, of philosophers who take Nietzsche seriously, especially in what he says
concerning knowledge, science, and truth’ (71). In another sense, the ‘postmodern’ part of
postmodern bourgeois liberalism is explicitly linked to Jean-François Lyotard’s apothegm
‘distrust of metanarratives’ (ORT, 199).
In the context of his views discussed in this chapter, Rorty’s postmodernism amounts to
the contention that we should give up the attempt to ground or justify our moral, social, and
political institutions on ahistorical, universalist, or metaphysical premises. Such premises have
taken different forms, from appeals to ‘intrinsic human dignity, intrinsic human rights, and an
ahistorical distinction between the demands of morality and prudence’ (ORT, 197), to the
universal possession of a faculty called Reason, or a Habermasian theory of intersubjective
communication. Such projects are, in Rorty’s eyes, problematic for several reasons. First of all,
appeals to human nature, or inherent moral worth are at odds with what might be understood as
the ethos of liberal democracies. Echoing his hero, American pragmatist John Dewey, Rorty held
that such absolutist notions are only helpful in so far ‘as they facilitate conversation about what
we should do next… [but] should be blurred or erased as soon as they begin to hinder such
conversation—to block the road of inquiry’ (ORT, 211). This is, however, exactly what such
notions do in practice. The attempt to justify the existence of political institutions on the basis of
something beyond time and chance is, for Rorty, precisely to take a step towards ending the
conversion about which laws should be in place, or which practices we should adopt or
abandon.8 Moreover, while Rorty himself takes these attempts to find a metaphysical
8 For Rorty, ‘A liberal society is one which is content to call “true” whatever the upshots of [a free and open
encounter] turns out to be. That is why a liberal society is badly served by an attempt to supply it with
11
underpinning for liberalism to be philosophically spurious, he thinks that they are losing
currency among intellectuals in Western culture more generally.9 Perhaps this is because two
thousand years of searching has failed to yield a consensus about what such an underpinning in
fact is. To borrow a term associated with the work of the philosopher of science Larry Lauden,
one might call this line of thought Rorty’s pessimistic induction against the need to
philosophically ground liberalism.
The second term of Rorty’s self-styled sobriquet (i.e., “bourgeois”) was meant to
emphasize the ‘Marxist claim that a lot of those [liberal] institutions and practices are possible
and justifiable only in certain historical and especially economic conditions’ (ORT, 198). The
institutions he has in mind include parliamentary democracy, procedural justice, or the freedom
of press. The practices include tolerance, the desire to resolve conflicts through discussion or
persuasion as opposed to force, to create a classless and casteless society in which individuals are
free to pursue their individual projects and private quests for self-fulfillment. The invocation of
appropriate historico-economic conditions is a reminder that such institutions and practices have
a history; they were the result of a great many historically contingent processes that have been
unfolding over the last two centuries—as opposed to the result of a discovery about human
nature, or the apprehension of an historical eschatology.10
Moreover, the continuation of such
institutions and practices depends, in part, on their propensity to augment the standard of living,
and degree of security for an increasing number of people.
“philosophical foundations.” For the attempt to supply such foundations presupposes a natural order of topics and
arguments which is prior to, and overrides the results of, encounters between old and new vocabularies’ (CIS, 52). 9 ‘As I see it, one important intellectual advance that has been made in our century is the steady decline in interest in
[the] quarrel between Plato and Nietzsche about what we are really like. There is a growing willingness to neglect
the question “What is our nature?” and to substitute the question “What can we make of ourselves?”’ (TP, 169). 10
In this respect, Rorty intends ‘bourgeois liberalism’ as a contrastive term with ‘philosophical liberalism.’
12
In Contingency, Irony, and Solidarity Rorty adopts Judith Shklar’s minimalist definition
of ‘liberal,’ as someone ‘who think[s] that cruelty is the worst thing we do’ (xv). I take it that this
is intended to rule out, on one hand, those who see nothing wrong with cruelty—especially when
directed at members of a group who are not deemed worthy of moral consideration. On the other
hand, this definition allows Rorty to rule out those willing to admit of possible transgressions
that lie beyond cruelty. For instance, by Shklar’s definition, a person who considers acting
against God’s will to be worse than acting cruelly would not get to count as a liberal.11
Rorty admits that the very idea of postmodern bourgeois liberalism has a paradoxical ring
to it. This is
partly because the majority of those who think of themselves as beyond metaphysics and
metanarratives also think of themselves as having opted out of the bourgeoisie. But partly
it is because it is hard to disentangle bourgeois liberal institutions from the vocabulary
that these institutions inherited from the Enlightenment (TP, 199).
This admission serves as the backdrop against which much of Rorty’s political writings are set.
On the one hand, it forces him to tell a story about how it is possible to jettison the metaphysical
underpinnings that have, historically, served to bolster liberal institutions, without having to
jettison those institutions themselves. On the other hand, he needs to persuade ‘the souls of
bourgeois liberals who have not yet gone postmodern’ (207)—i.e., those who are unwilling to
give up such underpinnings in the first place—that they are at best unnecessary and at worst
inimical to the continuation and improvement of such institutions.
The first, and perhaps crucial step that Rorty takes towards resolving this air of paradox is
to develop the—admittedly provocative—notion of ethnocentrism; an attempt to avoid what
11
This is not to say that Rorty’s account rules out other aspects of liberalism. Cf. ‘The point of liberal society
is…simply to make it as easy as possible for people to achieve their wildly different private ends without hurting
each other’ (EHO, 196).
13
Clifford Geertz calls ‘witless relativism or transcendental dogmatism’ (ORT, 203).
Ethnocentrism is a particularly thorny matter for postmodern bourgeois liberals like Rorty. As he
puts it:
When we [liberals] find ourselves reacting to the Nazis or the [religious] fundamentalists
with contempt—we have to think twice. For we are exemplifying the attitude we claim to
despise. We would rather die than be ethnocentric, but ethnocentrism is precisely the
conviction that one would rather die than share certain beliefs. We then find ourselves
wondering whether our own bourgeois liberalism is not just one more example of cultural
bias…[W]e begin to wonder whether our attempts to get other parts of the world to adopt
our culture are different in kind from the efforts of fundamentalist missionaries. If we
continue this line of thought too long we become “wet” liberals. We begin to lose any
capacity for moral indignation, any capacity to feel contempt… We have become so
open-minded that our brains have fallen out (203).
There is a sort of division of moral labour carried out by liberal democracies that, for Rorty,
makes them unique. It provides the basis for, if not a sort of ethnocentrism, then an anti-anti-
ethnocentrism with which their proponents justify contempt for Nazis or fundamentalists. He
writes that the ‘moral tasks of a liberal democracy are divided between the agents of love and the
agents of justice. In other words, such a democracy employs and empowers both connoisseurs of
diversity and guardians of universality’ (206). Agents of love include, but are not limited to
‘historians, novelists, ethnographers, and muckracking journalists’ (207). Their work serves to
expand the scope of moral consideration, through their insistence ‘that there are people out there
whom our society has failed to notice’ (206). Rorty thinks that patient and detailed
anthropological study functions in an importantly similar way as novels. Both types of projects,
he thinks, have the capacity to compel shifts of moral perspective in their audiences. They help
make what once seemed like irrationality, stupidity or sin, appear as simply unfamiliar. More
importantly, agents of love allow such unfamiliar beliefs, desires, and social practices to be seen
as morally irrelevant—they wield the most powerful tools which liberal democracies have
14
created for reducing cruelty and humiliation.12
Agents of justice are those who ‘make sure that
once… people are admitted as citizens, once they have been shepherded into the light by the
connoisseurs of diversity, they are treated like all the rest of us’ (206). I take it that, for Rorty,
the paradigmatic agents of justice are courts and legislators; but include activists of varying
stripes who work to improve procedural justice.
Rorty maintains that this division of moral labour is endemic to liberal democracies,
which is not to say that there is a necessary or essential connection between its existence and that
of such societies. Rather, the proliferation of agents of love and agents of justice is a contingent
historical byproduct—one for which Rorty is both grateful and proud. Moreover, his point is not
that these developments have created the ideal society—there is obviously plenty of work for
improvement; both in terms of people who have yet to receive the moral consideration they
ought to, as well as institutional failings that unfairly privilege certain social groups (despite the
de jure notion that things are otherwise). His point is that within this division of moral labour can
be found the seeds of liberal-democratic self-improvement. As a result, postmodern bourgeois
liberals can appeal to these concrete aspects of their society in order to make comparisons (albeit
invidious ones) between other social and cultural forms—past, present, and future. This anti-anti-
ethnocentrism allows Rorty to say at once that, in terms of social justice, liberal societies have
been making progress, and that they are, for example, better than the sort of society exemplified
by Nazi Germany. It allows him to do this without the help of a skyhook—without an appeal to
something like human nature, or intrinsic moral worth, which (for better or worse) have
historically been mainstays of liberal moral and political rhetoric.
12
Crucially, Rorty wants to contrast the effectiveness of such tools with argument. Moral progress is rarely (if ever)
the outcome of getting someone to agree that a certain conclusion follows from a set of true premises. He thinks that
we convince people to change their moral identities by getting them to read books and watch films that appeal to
their sentiments. For a more developed account of this claim see his essay entitled ‘Human Rights, Rationality, and
Sentimentality’ in TP.
15
This is a difficult balancing act, and Rorty is not without his critics. On the one hand, he
is accused of relativism by those who are unwilling to buy into his disavowal of metaphysics; to
relinquish the project of providing foundations for socio-political institutions. On the other hand,
Rorty’s commitment to liberal democratic institutions has led to charges that he is uncritically
apologetic for the status quo. He would spend the rest of his career attempting to assuage these
worries by doing what he knew how to do best: redescription. His 1989 book Contingency,
Irony, and Solidarity is a more fully-worked-out version of the metaphilosophical position just
outlined. The book is both an attempt to move beyond the traditional assumptions of political
philosophy, and to sketch the ideal inhabitant of Rorty’s postmodern bourgeois liberal society—
what he calls the liberal ironist.
Rorty offers a unique gloss on what can be understood as the traditional Western
philosophical approach to political philosophy which runs from Plato, through Kant, up to
contemporary thinkers like Jürgen Habermas. On this approach the aim of political theorizing is
to reconcile two seemingly irreconcilable desiderata: the striving for individual (or private)
fulfillment with social solidarity. For Rorty, the traditional attempt to ‘fuse’ these two
considerations within a theoretical framework, usually (if not always) gets off the ground
through an appeal to some sort of metaphysical principle.13
Kant, for example, spun a story about
how moral obligation (human solidarity) springs from the fact that human beings possess a
universal faculty of reason. Both Hobbes and Rousseau provided a theoretical justification for a
society in which individuals agree to relinquish their natural rights and uphold certain legal and
moral obligations to one another, on the basis of supposed insights (gleaned from imagining a
13
‘The attempt to fuse the public and private lies behind both Plato’s attempt to answer the question “Why is it in
one’s interest to be just?” and Christianity’s claim that perfect self-realization can be attained through service to
others’ (CIS, xiii).
16
supposed original state) about human nature. We have seen that Rorty shares the Nietzschean
suspicion of such attempts, and finds them inimical to the aims of democracies. At the same
time, we have seen that he still thinks it is worth preserving the very liberal social institutions
that arose from these attempts to fuse an individual’s private aims with her responsibilities to
others.
The response that Rorty offers to this politico-philosophical tradition is a thoroughly
pragmatic one. He suggests that we can compartmentalize our lives into a part where we—like
good Nietzscheans—pursue our own idiosyncratic quests for self-perfection, and another part in
which we fulfill our responsibilities to others (e.g., to reduce suffering and humiliation, to create
the casteless and classless society that was the vision of many Enlightenment thinkers). Rorty
maintains that there are certain writers who are exemplary at teaching us how to engage in
projects of self-perfection, and others who are especially insightful when it comes to discussing
our moral and social responsibilities. While he admits that, in principle, there might be some
extraordinary thinkers who are able to do both of these things, Rorty does not think that these
two activities (private and public philosophy) should be done at the same time.14
One way to
make sense of Rorty’s insistence on the public-private split, is to see it as stemming from his
linguisticism. Self-overcoming, or private perfection, is tantamount to coming up with new ways
of talking; it involves the quest for previously unfathomable self-descriptions, and especially
new ways of defining our relationship to the past. Jacques Derrida is, in Rorty’s eyes, the private
philosopher par excellence. Rorty writes that, in his later works
14
For Rorty, ‘Autonomy [i.e., private perfection] is not something which all human beings have within them and
which society can release by ceasing to repress them. It is something which certain particular human beings hope to
attain by self-creation, and which few actually do. The desire to be autonomous is not relevant to the liberal’s desire
to avoid cruelty and pain’ (CIS, 65).
17
Derrida privatizes his philosophical thinking, and thereby breaks down the tension
between ironism and theorizing. He simply drops theory—the attempt to see his
predecessors steadily and whole—in favor of fantasizing about those predecessors,
playing with them, giving them free rein to the trains of associations they produce. There
is no moral to these fantasies, nor any public (pedagogic or political) use to be made of
them; but, for Derrida’s readers, they may be nevertheless exemplary—suggestions of the
sort of thing one might do, a sort of thing rarely done before (CIS, 125).
Why does Rorty consider Derrida’s work unhelpful for political deliberation? Recall that the aim
of public philosophy is more or less coextensive with social justice—to clarify and enlarge the
scope of our moral community. In Rorty’s liberal utopia this ‘requires a banal moral
vocabulary—a vocabulary which is no more relevant to one’s individual private self-image than
to another’s. In a liberal society, our public dealings with our fellow citizens are not supposed to
be Romantic or inventive; they are supposed to have the routine intelligibility of the marketplace
or the courtroom’ (EHO, 196).
The ‘liberal ironist’ can be understood as the sort of person who has managed to
successfully carry out Rorty’s suggested compartmentalization between public and private. They
have been able to retain a sense of identity with and hope for liberal democratic institutions,
while at the same time fostering a sense of what Rorty calls ‘irony’ in a way that does not
interfere with loyalties to such institutions. Simply put, an ironist fully embraces the fact that the
language they use, their sense of self-hood, as well as the culture they inhabit, are all the results
of chance historical processes. Rorty’s ironist
faces up to the contingency of his or her own most central beliefs and desires—someone
sufficiently historicist and nominalist to have abandoned the idea that those central
beliefs and desires refer back to something beyond the reach of time and chance. Liberal
ironists are people who include among these ungroundable desires their own hope that
suffering will be diminished, that humiliation of human beings by other human beings
may cease (CIS, xv).
The figure of the liberal ironist gives Rorty a fresh way of responding to the paradox of
postmodern bourgeois liberalism, while at the same time carving out for himself a place in the
18
landscape of political philosophy. It represents, for instance, a middle-ground between two
figures with whom Rorty takes himself to have ‘wide areas of agreement’ but whose views differ
from his ‘in opposing ways’ (CIS, 61). He contends that ‘Michel Foucault is an ironist who is
unwilling to be a liberal, whereas Jürgen Habermas is a liberal who is unwilling to be an ironist’
(61). The latter shares Rorty’s commitment to liberal democracy, but still wants a theoretical
foundation on which to ground such a society. As Rorty puts it, their disagreement is ‘merely
philosophical’ (67), that is, he does not share Habermas’s desire ‘to “ground” democratic
institutions in the same way as Kant hoped to—but to do the job better, by invoking a notion of
“domination-free communication” to replace “respect for human dignity” as the aegis under
which society is to become more cosmopolitan and democratic’ (63). We have already seen
some of Rorty’s strategies for undermining such theoretical pretensions. For instance, his
insistence on the unsuitability of universalist, or absolutist notions for democratic conversations,
the analogy with the secularization of morality, and what might be thought of as his pessimistic
induction. By contrast, Rorty appears to take his disagreement with Foucault to be entirely
political. In particular, he frames the issue in terms of Foucault’s unwillingness to be a liberal. In
the next section I will attempt to specify what such a criticism is supposed to amount to, but
before doing so it is worth emphasizing the extent to which the two thinkers are in agreement.
As a fellow ‘ironist’ Foucault shares with Rorty the view that cultures, languages, and
senses of self are thoroughly contingent, and that we should give up the attempt to provide
theoretical justifications in advance for any set of social arrangements—liberal or otherwise.
Both thinkers agree with Nietzsche that the Truth—be it about human nature, or discoverable
within some timeless neutral vocabulary—will not set us free. And in practice, both Foucault and
Rorty would offer much of the same reasons against the project trying to seek philosophical
19
groundings for politics. This overlap is important, as it functions to set Rorty’s engagement apart
from many of Foucault’s liberal critics, who focus their critique on Foucault’s abandonment of
enlightenment notions of Truth and Reason. Indeed, dropping these notions has led both
Foucault and Rorty to be accused of espousing a dangerous form of relativism, often by the same
liberal critics. Thus, much of the criticisms made of both thinkers often result from the fact that
their interlocutors operate with radically different assumptions about, for example, the nature of
socio-political institutions or the importance of philosophical theory. Given that both Rorty and
Foucault appear to occupy so much of the same metaphilosophical ground, it is especially
worthwhile to clarify their political disagreement and to attempt to understand its source.
1.2 Rorty’s Disagreement with Foucault
What does it mean for someone to refuse to be a liberal? And why would anyone take
such a refusal to be problematic? Given Rorty’s penchant for repeating this criticism in several
different ways, it can be cashed out multifariously. The way I see it, Rorty’s central criticism of
Foucault amounts to a general disagreement which he takes to have two related, and equally
undesirable consequences. The central worry is that Foucault was ‘not willing to think of himself
as speaking as a member of any “we,”’ (CIS, 64), much less from the standpoint of the liberal
“we”. Rorty finds this problematic because he thinks that making meaningful social and political
contributions to a community requires that one (at least partially) view oneself as belonging to
that community. That is, to be able to engage in a conversation about the status of those
institutions and practices which help characterize that community, and to be able to propose
changes that might improve these institutions and practices, presupposes at least some overlap in
values, or shared aspirations. As a good pragmatist, it would be strange for Rorty to take issue
with Foucault’s refusal to see himself as a member of ‘we liberals,’ if he did not take it to make a
20
practical difference in Foucault’s political thought. So we must look to this refusal’s (perceived)
consequences. On the one hand, Rorty takes Foucault’s anti-liberalism to lead to a retrospective
refusal to admit that current social practices and institutions can be judged favorably in
comparison to former ones; and on the other, to a refusal to say how current social practices and
institutions could be improved. Rorty’s central criticism of Foucault is thus twofold;
unwillingness to admit of progress, and refusal to ‘sketch a Utopia.’
It is worth pointing to some of the places in Foucault’s writings from which Rorty derives
his criticisms. Often, he would associate the French thinker’s refusal to imagine better and
brighter societies with a revolutionary impulse resulting from the Marxist and Nietzschean
‘conviction that we are too far gone for reform work—that a convulsion is needed, that our
imagination and will are so limited by the socialization we have received that we are unable even
to propose an alternative to the society we have now’ (CIS, 64). Such a conviction was expressed
by Rorty’s “least favorite” thing that Foucault ever said: “I think that to imagine another system
is to extend our participation in the present system.”15
Indeed, Rorty’s deep repulsion for this
remark is itself evidenced by the fact that he would cite it at every conceivable opportunity (e.g.,
CIS, 64; PSH, 129). This remark, taken from a 1971 interview, chimes with another interview
(notably from the same year) called ‘On Popular Justice: A Discussion With Maoists,’ where
Foucault opposes the Maoist idea of people’s courts only on the basis of a suspicion that they
will simply take on many of the problematic elements of previous judicial system. He claims that
[t]he masses will discover a way of dealing with the problem of their enemies, of those
who individually or collectively have harmed them, methods of retribution which will
range from punishment to reeducation, without involving the form of the court which—in
any case in our society…is to be avoided. This is why I was against the people’s court as
15
Foucault’s remark is from a 1971 interview and can be found in Language, Counter-Memory, Practice: Selected
Essays and Interviews. Cornell University Press, 1980.
21
a solemn form, designed to synthesise, to replace, all other forms of struggle against the
judicial system. This seemed to me to re-legitimate a form which drags along with it too
much of the ideology imposed by the bourgeoisie… it is a dangerous weapon because it
will act as a precedent, and will be dangerous later on, within a revolutionary state
apparatus, because forms of legal proceedings will be subtly introduced into it which will
threaten to reestablish these divisions (PK, 28).
This passage exemplifies especially well both consequences of Foucault’s refusal to be a liberal
with which Rorty found issue. On the one hand, it seems to presuppose that the judicial apparatus
characteristic of Western societies is so problematic that it is beyond repair. A reformist liberal
such as Rorty would admit that the justice system (then and now) has its flaws, but would
presumably attempt to work within the existing political and juridical framework in order to
mitigate, and perhaps eventually eradicate such flaws. What motivates their confidence in such
an attempt? I take it that, for Rorty, part of the motivation stems from the belief in the
comparative superiority of these institutions to those of the past. That is, the belief that these
institutions are worth preserving and improving, because they are better than any comparable
previous institutions. When confronted with the above passage (and others like it), I take it that
Rorty interpreted Foucault’s unwillingness to even consider reforming the ‘judicial apparatus’ as
a token of his refusal to regard it as an instance of progress. That there is apparently nothing
worth preserving in these institutions suggests, to a liberal like Rorty, that there is simply no
story of progress to be told.
On the other hand, Foucault refuses all but a vague hint about what is to replace the
liberal judicial system. It is as though the very attempt to say in advance what should lie on the
other side of the revolution risks imbuing the future with the deleterious forms of the present.
Yet, if part of the reformist liberal’s confidence in the existing framework (i.e., to allow for the
eventual amelioration of institutions in particular and society more generally) stems from some
retrospective belief in progress, the rest of it stems from the hope they glean from concrete
22
proposals about how to bring such improvements about.16
Thus, Foucault’s unwillingness to
offer such concrete proposals was, on Rorty’s interpretation, just another consequence of his
reluctance to see himself as a liberal.
While I agree that (especially during the first few years of the 1970’s) there is sufficient
textual evidence to support the view that Foucault was unwilling to identify as a liberal—
precisely in the ‘double-refusal’ sense outlined above—whether he maintained this stance later in
his life is perhaps less clear. For now I pass over this question. Whether or not they reflect a sort
of radicalism that Foucault sustained throughout his career, these remarks form the locus of his
political disagreement with Rorty. Again, I take this disagreement to be unique in the sense that
those considerations available to many of Foucault’s other liberal critics are unavailable to Rorty.
For example, Rorty is unable to appeal to any philosophical (i.e., universal, ahistorical) basis on
which to convince Foucault to share his approbation of liberal societies. And while he may want
to accuse Foucault of leaving no room for hope in an emancipatory project, Rorty is not able to
advance such an accusation by appealing to Foucault’s skepticism about the notion of human
nature. As a result, Rorty seems to be left with very little dialectical space in which to frame the
tension he feels that Foucault creates for his own account. Barry Allen offers a helpful gloss on
this predicament by suggesting that Foucault puts a strain on Rorty’s distinction between public
and private:
[Foucault] seems to have chosen for his private poetry the assumptions of the public
institutions that Rorty most regards. He redescribes in imaginative new ways institutions
that Rorty does not want to have redescribed, because his loyalty towards them is not as it
were extensional, not independent of how they are described. If you say ‘modern
institutions of social welfare,’ Rorty might say, ‘It’s the right thing to do.’ But if the same
16
Consider this passage from Rorty: ‘We Deweyans [i.e., liberal pragmatists] have a story to tell about the progress
of our species, a story whose later episodes emphasize how things have been getting better in the West during the
last few centuries, and which concludes with some suggestions about how they might become better still in the next
few’ (ORT, 212).
23
history and rationality is redescribed in terms of biopower and a political anatomy of the
body, I think Rorty is unsure what that should imply for the solidarity he feels with
liberal institutions. His major criticism of Foucault is precisely that he doesn’t explain
anything on this point. He doesn’t tell ‘us’ what to do with the new language for
describing institutions to which ‘we’ feel loyal. Foucault may not feel that loyalty, or
perhaps chooses to keep it private. What may be mere discretion, Rorty reads as evasion,
ambivalence, and unwarranted skepticism’ (Allen, 2009, 68-9).
Implicit in this passage is that Rorty’s metaphilosophical stance towards politics leaves him
unable to leverage anything like an argument against Foucault’s apparent contempt for liberal
societies. If competing (re)descriptions are really what are at stake between the two thinkers,
how should we decide which to adopt? Rorty is, of course, happy to admit that Foucault’s
redescriptions of liberal institutions and societies in all their insidiousness, serve an
indispensable purpose—to show us how ‘today’s chains are often forged from the hammers that
struck off yesterday’s’ (TP, 320). When, however, it comes time to justify to Foucault the claim
that we need stories of liberal progress along with hopeful dreams of an even better liberal
future, he is more or less at a loss.
In a recent study, Wojciech Malecki (2011) takes up some of the interpretive strategies
that Rorty employed throughout his career in making sense of Foucault, and their political
disagreement. One of Malecki’s key insights is that Rorty’s primary strategy for critical
engagement with Foucault was to offer explanations or contextualization for their disagreements,
rather than to present arguments. That is, Rorty’s critical writings on Foucault were almost
always concerned with diagnosing their apparent political incongruity on the basis of (what for
lack of a better term) could be called ‘external’ factors, rather than as seeing it as stemming from
diverging philosophical commitments. Malecki dubs these diagnoses ‘sociological,’
‘psychological,’ and ‘historical’ (Malecki, 112).
24
As we have seen, there are two principle reasons that Rorty objects to Foucault’s political
stance. Summarizing the first of these, Malecki writes that,
What is probably most striking for philosophers such as Habermas, [Charles] Taylor, and
Rorty, is that being a perceptive historian as [Foucault] was, sensitive to all kinds of
“cunning ways”, in which the process of modernization brought with it new kinds of
oppression, he remained blind to the unquestionable advantages of that process and to
their “compensating” for the negative consequences of it. This seems all the more curious
if one considers the fact that it is precisely thanks to these advantages that Foucault could
fashion his life in a way that was undreamt of (or at least was fiercely punishable) before
modernity (110).
The second objection, is that Foucault ‘was unwilling to sketch a “utopia” ahead of us—both in
the sense of “liberal,” or “bourgeois,” utopia, which Rorty cared about the most…and any utopia
whatsoever’ (111). Yet unlike Foucault’s other liberal critics, Rorty preferred not to frame these
objections in terms of a philosophical difference between them (i.e., in terms of Foucault’s
relativism, or his anti-essentialism), but instead, sought to interpret their disagreement on the
basis of factors ‘external’ to their philosophical views.
This first type of explanation is manifest in an interview where Rorty chalks his
disagreement with Foucault up to ‘contingent French fashion.’17
According to Rorty ‘I don’t see
what Foucault had against bourgeois liberalism, except that in the France of the fifties and sixties
it just wasn’t respectable to be a bourgeois liberal.18
I don’t think he has any arguments against it
or anything better to suggest’ (TCF; quoted in Malecki, 112). Presumably, on this diagnosis, had
17
It is worth noting that Malecki suggests that the term ‘sociological’ in this case should ‘be qualified with the
adverb ‘vulgarly’, since it boils down to a sweeping claim that Foucault’s behavior is the result of following some
“contingent French fashion.” (112). 18
Rorty appears to overlook the possibility of much deeper historical reasons for the contempt which many French
intellectuals had (and perhaps still have) for liberalism. One explanation involves the failure of both the French
Third and Fourth Republics: The former ending in the Nazi occupation of France in the 1940’s—giving rise to the
so-called Vichy government, and the latter (though initially coincident with economic development and important
social reform) leading to a series of crises involving decolonization (namely the Algerian war of 1958). A second
(and perhaps concomitant) explanation is the anti-materialism and contempt for bourgeois culture in French culture
that can be traced back to the French Revolution. I am indebted to Bruce Hunter for suggesting both of these
explanations.
25
Foucault grown up in America rather than in France, he would have been more sympathetic to
the ‘we liberals’ with whom Rorty wanted him to identify. This gives rise to the question: what if
Rorty had grown up in France, and attended the École Normale Supérieure; become friends with
Maoists, and experienced the aftermath of May 1968? Would he still have had the same affection
for bourgeois liberal institutions? Perhaps not, and by Rorty’s own account. As Malecki and
others point out, Rorty did not hesitate to tether his biography to his political stance. His parents
were at the fringe of the so-called New York Intellectuals of the early 1930’s—the group
consisted of the first wave of leftists to break from the American communist party and openly
criticize communism under Stalin. As a result, Rorty grew up, as Malecki puts it, ‘a liberal who
would squirm with revulsion at each positive mention of Stalin (or any other communist leader)
and at each justification of terror in leftist politics’ (113). I do not think that Rorty’s invocation
of autobiographical matters was ever meant to be an apology or justification for his die-hard
liberalism. If anything, I take them as reminders of the seriousness with which he approached the
idea, shared by Foucault, that one’s political beliefs are historically and socially contingent.
Along with this attempt to provide a sociological contextualization of their apparent
political differences, Malecki contends that in some cases, Rorty would appeal to either a
“psychological” or “historical” explanation. These are closely related insofar as they amount to a
rejection of liberal societies on the basis of the hopes of encountering some previously unrealized
forms of subjectivity, or of recovering a long-forgotten historical one. In the first case, Rorty
construes Foucault’s refusal to buy into liberalism as the result of a Nietzschean longing to
invent or discover a new type of individuality (and its associated values) on the basis of which
one might reorganize political aims. As Malecki puts it, he interprets Foucault as having looked
‘contemptuously at liberal democracies from the standpoint of some further unspecified future
26
world that knows no happiness and whose contours he caught a glimpse of (through a glass
darkly, but nevertheless did) by dint of his experiences with sadomasochism and drugs’
(Malecki, 116).19
In the second case, Foucault’s historical research is supposed to have provided
him with insights into forms of subjectivity that can help us to shed those problematic ones that
have been produced by modernity.20
As evidence for this sort of reading, Rorty would frequently cite passages such as the
following, taken from the afterword Foucault had written for Dreyfus and Rabinow’s Michel
Foucault: Beyond Structuralism and Hermeneutics:
Maybe the target nowadays is not to discover what we are, but to refuse what we are. We
have to imagine and to build up what we could be to get rid of this political “double
bind,” which is the simultaneous individualization and totalization of modern power
structures. The conclusion would be that the political, ethical, social, philosophical
problems of our days is not to try to liberate the individual from the state, and from the
state’s institutions, but to liberate us both from the state and from the type of
individualization which is linked to the state. We have to promote new forms of
subjectivity through the refusal of this kind of individuality which has been imposed on
us for several centuries (Foucault, ‘The Subject and Power’, 216).
Ultimately, Rorty seems to have gravitated towards the “psychological” explanation of his
political disagreement with Foucault. This comes across most apparently in his review of James
Miller’s popular but controversial biography entitled The Passion of Michel Foucault, where
Rorty notes an ambiguity in Foucault’s thought that, no doubt, he takes to be central to their
disagreement. He writes that,
19
I take it that (at least before having read James Miller’s biography of Foucault), Rorty would have had the
following sort of passage in mind when attributing this sort of view to Foucault: ‘It is possible that the rough outline
of a future society is supplied by the recent experiences with drugs, sex, communes, other forms of consciousness,
and other forms of individuality. If scientific socialism emerged from the Utopias of the nineteenth century, it is
possible that a real socialization will emerge, in the twentieth century, from experiences’ (Language, Counter-
Memory, Practice, 231). 20
As Rorty notes, Foucault explicitly repudiates this sort of interpretation. See for example, when asked whether he
thought that the Ancient Greek ethical approach to sex offered ‘an attractive and plausible alternative’ to our own,
Foucault responded by saying, ‘No! I am not looking for an alternative; you can’t find the solution of a problem in
the solution of another problem raised at another moment by another people’ (FR, 343).
27
Instead of sketching an alternative utopia, Foucault keeps hinting that if we could only
break out of the meshes of power which have formed our subjectivity, we would
appreciate how contemptible the last men are, and would realize that our “humanist”
desire for shared happiness is not only an artifact of acculturation, but one which we
should repudiate. But he never, as far as I can see, makes clear whether he thinks that (a)
there is something deep inside us which is repressed by any socialization, subjectivity-
forming, process, and nevertheless needs to be freed, or (b) that prior to the modern state
we were socialized into a different, better, sort of subjectivity than we have now, or (c)
that the sorts of “limit experiences” he found in LSD trips and in gay bathhouses are a
token of the existence of human possibilities which no society has yet explored, but
which might somehow become the basis for a new, non-humanist, politics (64-5).
After having read Miller’s book and noting that Foucault would often explicitly reject the first
two possibilities, Rorty is left convinced that the third is the most likely.
Setting aside the question of whether or not Rorty simply misread Foucault, I want to
emphasise that Malecki’s study demonstrates Rorty’s reluctance to see (what he took to be)
Foucault’s radicalism as flowing out of any substantive philosophical commitment. That Rorty
continuously played up these ‘sociological’ ‘historical’ and ‘psychological’ interpretive
strategies echoes the claim that his disagreement with Foucault was indeed political, rather than
‘merely philosophical’. I shall attempt to cast doubt on this view in the next chapter.
28
Chapter Two: Conceptual Relativism
So far I have offered an account of the political disagreement between Rorty and
Foucault. Rorty is a professed liberal who believes that progress can be made by piecemeal
reform of current institutions and social practices. He chides Foucault’s unwillingness to identify
as a liberal, which for Rorty, amounts to a reluctance to praise current liberal societies in
comparison to the past, as well as a refusal to sketch a future utopia. As Wojciech Malecki has
illustrated, Rorty adopted different strategies for diagnosing this political disagreement,
preferring to impute ‘external’ (i.e., biographical and cultural) differences as its locus. That is, at
least by Rorty’s own lights, there is no philosophical reason that can explain why he is
committed to reformist liberalism, and Foucault to a form of radicalism.
In this chapter I argue that there is a substantive, and overlooked philosophical difference
between the two thinkers, which can, moreover, explain their political divergence. This
philosophical issue in question is one that holds an important, but occasionally overworked place
in twentieth-century Anglophone philosophy, but which has been encountered by a broad range
of thinkers across many traditions, and in various forms: the problem of incommensurability.
I shall begin by discussing one of the very first things that Rorty ever wrote about
Foucault: an essay entitled, ‘Foucault and Epistemology’ in which he attributes to the French
thinker a radical Nietzschean view about history, which gives rise to a sort of
‘incommensurability’ between vocabularies. This article is significant because it is an anomaly.
Rorty would never again attribute this view to Foucault. When the article was re-published a few
years later in an anthology (Foucault: A Critical Reader), Rorty admitted that his understanding
29
of Foucault had since changed.21
In particular, he appears to have given up on attributing to
Foucault the view that discourses inhabiting different historical epochs could be characterized as
incommensurable.22
Another reason the article is an anomaly is because Rorty refrains from
making what, for him, should be a patently obvious criticism of Foucault. For the better part of a
decade, Rorty had been championing a thesis advanced by the philosopher Donald Davidson,
which holds that if incommensurability (i.e., between languages, or conceptual schemes) is
understood to mean something like ‘untranslatability between languages,’ it is at best wildly
implausible; at worst, completely incoherent. Why did Rorty let Foucault off the hook? Only a
few years later he would premise a response to Jean-François Lyotard, almost entirely on the
basis of the latter’s commitment to the notion that there are so many incommensurable islets of
language. Perhaps this speaks to Rorty’s reverence for Foucault: he did not want to attribute to
the French thinker a philosophical view that he, himself found to be so patently false. I take this
to show that the problem of incommensurability (also known as conceptual relativism) plays an
ambiguous role in Rorty’s thought. On the one hand, he didn’t think that philosophy had much to
offer to politics. Repeatedly, he would insist on the ‘priority of democracy to philosophy,’ and
expressed hope that leftist intellectuals would ‘kick their philosophy habit’.23
On the other hand,
in his ‘Reply to Lyotard’ Rorty seems to premise his liberalism on the philosophical claim that
the very idea of incommensurable languages is unintelligible.
21
Although he did not specify what he had changed his mind about. See footnote (pp. 41). 22
In Chapter Three, I shall argue that Foucault’s commitment to conceptual relativism can help explain the political
disagreement between him and Rorty. Although my argument does not hinge on whether or not Rorty attributed to
Foucault specific views about incommensurability, my motivation for discussing Rorty’s claims in ‘Foucault and
Epistemology’ is that they provide a helpful starting point for interrogating both thinkers views regarding
incommensurability. 23
See for example ‘The Priority of Democracy to Philosophy’ in ORT (pp. 175-96); and Achieving our Country
(especially section three entitled, ‘A Cultural Left).
30
I aim to show that Rorty’s initial intuitions were correct; that Foucault did in fact
subscribe to some form of conceptual relativism, at least in his so-called archaeological writings.
Fleshing out the role that incommensurability plays in Foucault’s work will provide a backdrop
for the next chapter where I discuss the relationship between this philosophical question, and
Rorty and Foucault’s political attitudes.
2.1 Foucault and Epistemology
In ‘Foucault and Epistemology’ Rorty urges philosophers (and perhaps Foucault himself)
not to look for a theory of knowledge in the pages of The Order of Things or the Archaeology of
Knowledge (41). He begins by suggesting that there are two ways of dealing with the familiar
Foucauldian insight that ‘objects constitute themselves in discourse’. One might suggest—as Ian
Hacking does—that such a remark exemplifies a ‘central topic of the theory of knowledge’
(Rorty, 42). Alternatively, one might take a ‘deflationary’ stance towards Foucault’s insight.
Everybody knows that the way in which people use language can bring ‘objects’ into being, ‘in
the sense that there are a lot of things which wouldn’t exist unless people had come to talk in
certain ways’ (42). For the deflationist, there is nothing philosophically mysterious about the fact
that we would not have particular academic disciplines or governmental institutions unless
particular socio-linguistic practices were in place. Intellectual history can shed light on how we
came to have such practices, but not epistemology.
Rorty thinks that Foucault is torn. In some passages the French thinker ‘wants to give up
all the traditional notions which made up the “system of possibilities” of a theory of knowledge’
(43), whereas in other passages he is unwilling to take the deflationary stance toward his own
remarks. That is, ‘he is not content simply to give a genealogy of epistemology… [r]ather he
31
wants to do something like epistemology’ (43). To bring this tension into focus, Rorty sketches
three different ‘attitudes’ that a thinker might adopt towards the very notion of a theory of
knowledge. The ‘Cartesian’ attitude begins with the assumption that culture is and should be
compartmentalized hierarchically. There are areas—paradigmatically the natural sciences—
where objectivity and rationality reside, and other areas such as the arts or religion where we
should not expect to find such virtues. Looking to these former areas of culture, according to the
Cartesian, will provide us with insight into the nature of knowledge. Typically, according to
Rorty, this nature gets cashed out in terms of ‘representation’ or ‘correspondence’. That is to say,
only those general relations between human mental states or statements which accurately
represent or correspond to their objects, or states of affairs, etc., are the ones that count as
knowledge. By contrast, the ‘Hegelian’ attitude ‘takes for granted that rationality is to be viewed
sociologically and historically’ (44). It eschews notions like ‘correspondence’ or ‘accuracy of
representation’ in favor of the view that intellectual progress amounts to convergence at the end
of inquiry. Rather than compartmentalize culture into those areas which exhibit rationality and
objectivity and those that do not, the ‘Hegelian’ ‘sees history as doing what the Cartesian thought
epistemology should do—exhibiting the superiority of the present to the past, and giving helpful
hints to backward areas of culture as to how they might catch up with the more progressive
areas’ (45).
Rorty admits to having attributed (mistakenly) to Foucault the Hegelian attitude toward
epistemology, upon first having read him. Steeped in the intellectual environment of post-war
Anglo-American philosophy departments, Rorty once believed, naïvely, that historicism was
about as radical a doctrine as one could hope for. To his American ears ‘the suggestion that there
is no such thing as an ahistorical nature of knowledge or of rationality to be discovered by
32
philosophical analysis is so titillating that [he assumed] that Foucault must be getting the same
kick out of it’ (45). To make this assumption was to fail to appreciate the impact that Nietzsche
had on Foucault. Not only is Hegelian historicism ‘old hat’ on the post-Nietzschean continent, as
Rorty came to realize, but Foucault sees it as having the same shortcoming as the Cartesian
attitude. That is, both accounts share the
conviction that there is a way of rising above the present and viewing it in relation to
inquiry in general. The Cartesian does this by discovering the ahistorical nature of
rational inquiry. The Hegelian does it historically by contrasting the present state of
inquiry with the Peircian convergence towards the true and the real which we would
expect given both the ‘ideal speech situation’ and unlimited grant money (46).
It is precisely the attempt—common to both the Cartesian and the Hegelian epistemological
attitudes—to glean an account of intellectual or historical progress by stepping outside of
existing social practices that Rorty takes Foucault to be rejecting. The assumption that objectivity
can be cashed out as either increasingly accurate representations of mind independent reality, or
though Hegelian historical eschatology, are for Nietzsche and Foucault promissory notes that we
ought not to accept. On Rorty’s reading, Foucault’s insistence that the elements of chance,
contingency, and ruptures be at the center of historiography reflects his willingness to take
seriously the idea that ‘the vocabulary any community—even an ideal one—uses is just one
more vocabulary, and may be as incommensurable with its predecessors as ours with
Paracelsus’s’ (48, italics added). Full appreciation of this Nietzschean attitude, Rorty contends,
leads one to suspect that any attempt to tell a story of intellectual, moral, or social progress will
be too ‘Whiggish’ or ‘anachronistic’ to be of any use. Hence, Foucault’s attempt to ‘write history
in a way which will destroy the notion of historical progress’ (46).
Having presented this rough sketch, Rorty finds it difficult to see how Foucault could
even be in a position to offer a successor subject to epistemology, or even a ‘theory of anything’
33
(47). But if Foucault’s Nietzschean attitude prevents him from engaging in any theoretical
enterprise, what exactly do his (pre-1970’s) books accomplish? Rorty suggests that
[a]s far as I can see, all [Foucault] has to offer are brilliant redescriptions of the past,
supplemented by helpful hints on how to avoid being trapped by old historiographical
assumptions. These hints consist largely in saying: do not look for progress or meaning in
history; do not see the history of a given activity, of any segment of culture, as the
development of rationality or of freedom; do not use any philosophical vocabulary to
characterize the essence of such activity or the goal it serves; do not assume that the way
this activity is presently conducted gives any clue to the goals it served in the past (47).
These hints are entirely negative, and should not be understood as the product of a philosophical
theory (or as an outline for a general method for doing history). For present purposes, what is
worth noting is that at this early stage in his career, Rorty refrains from offering any overt
criticism of Foucault—or of the ‘Nietzschean’ attitude toward epistemology in general. Rather,
he expresses a cautious ambivalence towards the possibility of a culture where such ideas would
be the norm. That is, a culture which refused to tell itself any stories of intellectual, social, or
moral progress—one ‘that had genealogies, but no eschatology’ (48). Rorty can hardly fathom
what it would be like to inhabit such a culture, but concludes his essay by expressing gratitude
towards Foucault for attempting to convey what it might be like for humans to relinquish all
forms of ‘metaphysical comfort’.
Roughly a year later, Rorty undertook another discussion of Foucault in ‘Method, Social
Science, and Social Hope’. This essay presents some of Rorty’s most developed insights
regarding the debate between proponents of the so-called ‘hermeneutic’ approach to social
science, and those advocating a ‘value-free’ approach (i.e., that which attempts to model the
social sciences on the natural sciences in a way that is quantitative and often behavioristic). Here
his discussion of Foucault is set against the backdrop of the following two pragmatist views
about the social sciences. The first is that once one abandons the ideal of a nature or essence
34
common to all human beings, there is no reason to think that the human sciences will themselves
have a nature which requires a special methodology. Therefore one will not feel the need to
debate whether the social sciences should be ‘value-free’ or ‘hermeneutical’ in method. Second,
renouncing the traditional notion of ‘knowledge as accurate representation’ means that there is
no point in compartmentalizing culture into ‘hard objective’ areas as opposed to ‘soft subjective’
ones. In other words, we can see the disciplinary boundaries between natural and social sciences
as being drawn not by ontological considerations, but simply by practical concerns.
Once these pragmatic considerations are adopted, Rorty thinks that there are still
(roughly) two different stances that one might take toward the status of social sciences. The
first—exemplified by John Dewey—emphasizes their moral significance. By offering accounts
which allow unfamiliar social practices to be viewed as familiar, social scientists perform the
important function of expanding our moral community.24
By contrast, the second stance, which
Rorty attributes to Foucault, is one that emphasizes their negative, insidious outcomes. The latter
stance shows the ‘dark side’ of the social sciences insofar as they have served as ‘instruments of
the “disciplinary society”’ (CP, 204). If the Deweyan outlook plays up the connection between
knowledge and human solidarity, the Foucauldian outlook turns our attention to the often
overlooked connection between knowledge and power. Nonetheless, Rorty thinks that ‘[w]e
should see Dewey and Foucault as differing not over a theoretical issue, but over what we may
hope’ (204). He contends that both thinkers
make exactly the same criticism of the [philosophical] tradition. They agree, right down
the line, about the need to abandon traditional notions of rationality, objectivity, method,
and truth. They are both, so to speak, “beyond method.” They agree that rationality is
what history and society make it—that there is no overarching ahistorical structure (the
Nature of Man, the laws of human behavior, the Moral Law, the Nature of Society) to be
24
Recall the discussion of ‘agents of love’ in Chapter One.
35
discovered. They share the Whewellian and Kuhnian notion of Galilean science—as
exemplifying the power of new vocabularies rather than offering the secret of scientific
success. But Dewey emphasizes that this move “beyond method” gives mankind the
opportunity to grow up, to be free to make itself, rather than seeking direction from some
imagined outside source…Foucault also moves beyond the traditional ideals of method
and rationality as antecedent constraints upon inquiry, but he views this move as the
Nietzschean realization that all knowledge-claims are moves in a power-game (205).
If there is no theoretical difference between Dewey and Foucault, how should we make sense of
their apparent disagreement over the status of the social sciences? Moreover, how is Rorty to
justify his preference for Dewey’s hopeful disposition? Unlike Foucault, Rorty thinks that
Dewey took pains to ensure that the vocabulary (i.e., the rhetoric of ‘American pluralism’) which
he put forth to replace that of enlightenment rationalism, left open the possibility of ‘unjustifiable
hope, and an ungroundable but vital sense of human solidarity’ (208). It is precisely for this
reason that Rorty voices his preference for Dewey over Foucault. In an essay written at roughly
the same time Rorty contrasts the pragmatism he reads in thinkers like William James and
Harold Bloom, with that of Nietzsche and Foucault. He writes that
[p]ragmatism appears in James and Bloom as an identification with the struggles of finite
men. In Foucault and Nietzsche it appears as contempt for one’s own finitude, as a search
for some mighty, inhuman force to which one can yield up one’s identity. Bloom’s way
of dealing with texts preserves our sense of a common human finitude by moving back
and forth between poet and his poem. Foucault’s way of dealing with texts is designed to
eliminate the author—and indeed the very idea of “man”—altogether. I have no wish to
defend Foucault’s inhumanism, and every wish to praise Bloom’s sense of our common
human lot. But I do not know how to back up this preference with argument, or even with
a precise account of the relevant differences. To do so would involve a full-scale
discussion of the possibility of combining private fulfillment, self-realization, with public
morality, a concern for justice (CP, 158, my italics).25
There are two aspects worth noting about this early juncture in Rorty’s dealings with Foucault.
The first is that the passage just quoted contains the seeds of the central problematic that Rorty
25
The essay is called “Nineteenth-Century Idealism and Twentieth-Century Textualism.” It ‘was written for a
conference honoring Maurice Mandelbaum… in 1980’ (CP, x-xi). It originally was published in the Monist in 1981,
and was reprinted in Consequences of Pragmatism (Essays: 1972-1980).
36
would eventually take up nearly a decade later in Contingency, Irony, and Solidarity.26
It would
seem that as early as 1980, Rorty was troubled by the fact that he was unable to offer reasons for
his preference for the possibility of ‘unjustifiable social hope’ and ‘ungroundable sense of
solidarity’ left open in Dewey’s pragmatism (as opposed to Foucault’s pessimistic Nietzschean
pragmatism). The fact that he would spend the better part of the next decade formulating and
refining such a full-scale discussion of the possibility of combining private fulfillment with
public morality is a testament to the seriousness of the challenge that Foucault’s work presented
for Rorty.
The second point is that despite Rorty’s insistence that there is no theoretical difference
between Foucault and Dewey (and thus no argument that would recommend the latter’s
optimism), when read against the central claims outlined above in ‘Foucault and Epistemology’
it seems that there is such a disagreement lurking in the background. Whereas ‘Method, Social
Science, and Social Hope’ presents the difference between Dewey’s unjustifiable hope and
Foucault’s pessimism as almost a matter of personal taste, in the former essay, Rorty does in fact
offer what looks to be a theoretical difference between the ‘Hegelian’ and ‘Nietzschean’ attitudes
towards epistemology. Recall that the main contention was that while the former was willing to
tell stories of moral or intellectual progress, the latter viewed such stories with deep suspicion.
But Rorty took this deep suspicion to be the conclusion of something like the following
argument: if we accept the radical contingency of history (i.e., that it is permeated with chance,
ruptures etc.), then we must accept that any vocabulary we happen to use (i.e., moral, scientific,
political) is itself radically contingent and may be incommensurable with those vocabularies
employed by different epochs. But if we accept that our current vocabularies may be
26
See Chapter One for my account of Rorty’s public-private split.
37
incommensurable with those that precede and succeed them, then any attempt at telling a story of
(e.g., moral or scientific) progress will be too Whiggish or anachronistic to be of any use. Should
Foucault actually be committed to this line of reasoning, it seems that there would be a
theoretical disagreement between him and Dewey (and thus Rorty), insofar as the former’s
understanding of history would preclude him from advancing the sort of social hope that Rorty
admires in the latter. The fact that Rorty seems to have set this argument aside by the time he
undertook a comparison of Dewey and Foucault is suggestive. Perhaps he became reluctant to
ascribe such a view to Foucault—especially in light of his later writings. Whatever Rorty’s
reasons in fact were, it is surprising that in the paper he lets the notion of ‘incommensurability’
go unchallenged.
2.2 Rorty and the Davidsonian Attack on Conceptual Relativism
In Philosophy and the Mirror of Nature, Rorty’s account of incommensurability plays a
central role in his attempt to undermine what he takes to be philosophy’s self-conception, which,
as I understand it, consists of two dimensions. The first involves the idea that philosophy treats
sets of perennial, unchanging problems. In this sense, Rorty appears to establish a sort of
incommensurability between different philosophical epochs by playing up shifts in methodology,
conceptual changes, as well as transformations in the criteria which guide inquiry by defining
problems and determining what counts as an adequate solution. The second dimension involves
the idea that philosophers occupy a privileged role with respect to the rest of culture. On Rorty’s
reading of the Western philosophical tradition, the goal of epistemology is to provide a neutral
framework that will ensure commensurability between all ‘rational’ or ‘cognitive’ discourse
(PMN, 316). Given that his book can be read as an attempt to undermine this very project, Rorty
38
can also be thought of as opposing the idea of ‘commensurability’ in this second sense.27
That is,
however, not to say that he endorses conceptual relativism.
In both cases, Rorty’s attack on philosophy’s self-conception draws from the writings of
the physicist, and historian of science, T. S. Kuhn. In his now classic book The Structure of
Scientific Revolutions, Kuhn sought to undermine the notion that science is a ‘constellation of
facts, theories, and methods’ inhering in scientific textbooks, whereby scientific achievements
are viewed as additions to an ever-growing repository of scientific knowledge. Rather, Kuhn
suggested that history shows scientific activity to proceed through cycles rather than as a linear,
cumulative developmental process. In periods of what Kuhn calls ‘normal science’, researchers
proceed on the basis of shared acceptance of a ‘paradigm’—one or more examples of successful
research that ‘define the legitimate problems and methods of a research field for succeeding
generations of practitioners’ (10). Scientific paradigms attract researchers on the basis of their
novelty, and their propensity to give rise to a substantial body of problems that will guide future
research. During periods of normal science, foundational or theoretical questions about the
paradigms themselves do not arise, and scientists seldom attempt to produce novel discoveries.
Instead, by presupposing that the problems defined by the paradigm have a solution, scientists
behave like ‘puzzle-solvers’ in precisely the same sense as those who attempt crosswords or
jigsaw puzzles (36). Eventually, Kuhn contends, normal research will encounter a body of
recalcitrant problems, or anomalies, that cannot be accounted for by the presuppositions of the
paradigm. In some cases, should a significant number of these anomalies accrue, paradigms may
undergo revision; but in other cases, scientists may be led to develop and propose novel theories,
intended to take the place of the dominant paradigm. Kuhn calls this latter case a scientific
27
Rorty suggests that for contributions to a discourse to be commensurable means that they can be subsumed under
a set of rules which would determine how rational agreement between them is to be reached.
39
revolution, or a ‘non-cumulative developmental episode in which an older paradigm is replaced
in whole or in part by an incompatible new one’ (92). While like Rorty, Kuhn argues against the
availability of a set of antecedent criteria to serve as an algorithm for determining which
paradigm it is rational to accept,28
in certain passages he appears to go a step further, arguing that
competing scientific paradigms are so incommensurable that when a researcher abandons one for
another, she or he undergoes a sort of gestalt shift. As Rorty notes, in some of his more
‘idealistic-sounding’ moments, Kuhn made it seem as though scientists operating under different
paradigms ‘see different things’ or inhabit ‘different worlds’ (PMN, 324).29
Rorty is happy to apply Kuhn’s insights regarding theory change to the task of calling
into question both the understanding of philosophy as dealing with an unchanging set of
problems, as well as the epistemological project of seeking a neutral framework. He is, however,
unwilling to endorse the stronger ‘idealistic-sounding’ claims that Kuhn makes about researchers
employing different paradigms. The main reason for this has to do with Rorty’s seemingly
unshakable commitment to the work of Donald Davidson, whose attack on the scheme-content
distinction, or ‘third dogma of empiricism’ aims to undermine the very notion of a conceptual
scheme (and a fortiori the notion of incommensurable conceptual schemes, or languages etc.).30
28
Rather, Kuhn suggested that scientists share a set of values (i.e., simplicity, scope, explanatory power, etc.) to
which they appeal in choosing among competing theories. While such values, being shared, are objective in an
important sense, their application is always subjective (i.e., relative to an individual scientist’s background, interests,
etc.). 29
See especially chapter 10 of Kuhn’s The Structure of Scientific Revolutions. Kuhn contends that ‘at times of
[scientific] revolution, when the normal-scientific tradition changes, the scientist’s perception of his environment
must be re-educated—in some familiar situation he must learn to see a new gestalt. After he has done so the world
of his research will seem, here and there, incommensurable with the one he had inhabited before’ (112). 30
As I read Davidson, he develops two related, but distinct strategies for undermining the notion of
incommensurability. The first stems from his work on interpretation, and the second involves an attempt to call into
question the picture—inherited from Kant—of language, or thought as a conceptual scheme that ‘fits’ or ‘organizes’
something (such as ‘the world’, or ‘experience’). As I take his first strategy to pose the most obvious challenge to
the sort of incommensurability found in Foucault (discussed below), I will limit my discussion to it.
40
The notion of incommensurable conceptual schemes can be cashed out in various ways.
One might proceed by positing a belief system that is completely outlandish—which is, from our
point of view—largely false and irrational. Or, one might posit a belief system that is simply
incomprehensible by our lights. In the first case, we encounter a breakdown in shared standards
of rationality (raising the question of which standards we should adopt), but in the second case,
we aren’t even able to glean what standards of rationality are being employed (let along figure
out what utterances mean or beliefs are about). Davidson offers an account of interpretation that
is supposed to make either scenario seem spurious.
For Davidson, there is an inextricable connection between the attribution of language use,
and the attribution of intentional attitudes. As he puts it, while ‘speech requires a multitude of
finely discriminated intentions and beliefs’ (1973-4, 8), it is implausible that we can attribute
such complex attitudes to a speaker without being able to interpret what her words mean (which
is to say, to be able to translate her language into ours). In other words, we are unable to discover
what the user of another language means by her utterances, without discovering (or assuming) a
great deal about what she believes, and vice versa. To appreciate this point, Davidson would
have us consider the task of a field linguist interpreting the speech of a completely foreign
language user. Following his teacher W.V.O. Quine, he takes as basic certain general attitudes
which the imagined foreign language user would exhibit toward sentences such as ‘taking to be
true’. This provides the starting point for the linguist to construct a translation manual, which
correlates the speakers utterances with the circumstances in which they are uttered (i.e., their
truth conditions). Davidson contends that
if all we know is what sentences a speaker holds true, and we cannot assume his language
is our own, then we cannot take even a first step towards interpretation without knowing
or assuming a great deal about the speaker’s beliefs. Since knowledge of beliefs comes
41
only with the ability to interpret words, the only possibility at the start is to assume
general agreement on beliefs (18-9).
This ‘principle of charity’ is, for Davidson, the only non-arbitrary means available for the field
linguist to interpret the foreign language user’s speech. It requires attributing mostly true beliefs
to the foreign language user, as well as shared standards of rationality. That is, the field linguist
must read his or her logic into speaker’s beliefs. Both dimensions of the principle of charity have
consequences for both varieties of imagined incommensurable conceptual schemes mentioned
above.
In the first scenario—where we encounter an outlandish, and mostly false belief
system—the principle of charity suggests that the very possibility of establishing that a foreign
language user holds any given belief that is different than ours, or uses a concept for which there
is no analogue in our own language, presupposes a widespread basis of shared belief and
meaning. But if this is correct, there doesn’t seem to be much sense in calling such differences
incommensurable. Regarding the second scenario—where it is supposed that the contents and
logic of a belief system are inaccessible from our standpoint—Davidson’s account suggests that
we would have no more reason to draw this conclusion, than to conclude that we were faced with
meaningless marks or noises.
Rorty summarizes these insights quite nicely. Unless one construes ‘meanings’ as mental
essences or ideas in the minds of language users, ‘one will not be able to draw a clear distinction
between [a foreign speaker] using words differently in meaning from any words in our language
and [their] having many false beliefs’ (CP, 6). Of course there may be particular instances where
divergent beliefs make translation of a particular utterance tenuous, but we would be inclined to
view any purported translation manual that entailed that all of the foreign language user’s beliefs
42
about the world were false, as merely an unsuccessful translation manual. Moreover, taking
Davidson’s argument a step further, Rorty suggests that in a situation where we would be
inclined to throw up our hands and say that translation is impossible, we would have no reason to
suppose we are even encountering language users. It is, he thinks,
possible to imagine humanoid organisms making sounds of great variety at one another in
various circumstances with what appear to be various effects upon the interlocutors’
behavior. But suppose that repeated attempts systematically to correlate these sounds
with the organisms’ environment and behavior fail. What should we say? One suggestion
might be that the analytic hypotheses we are using in our tentative translation schemes
use concepts that we do not share with the natives… But could there be a way of deciding
between this suggestion and the possibility that the organisms’ sounds are just sounds?
Once we imagine different ways of carving up the world, nothing could stop us from
attributing “untranslatable languages” to anything that emits a variety of signals. But, so
this verificationist argument concludes, this degree of open-endedness shows us that the
purported notion of an untranslatable language is as fanciful as that of an invisible color
(CP, 6).
This last point raises an important question about the status of Davidson’s argument against
conceptual relativism (especially as it is conceived by Rorty). On the one hand, it is possible to
read Davidson as making a transcendental point about thought and language—whereby the very
notion of radically different conceptual schemes or languages is a priori impossible, or
incomprehensible. On the other hand, he can be read as making a much more pragmatic (and
perhaps negative) point. That given enough time, patience and willingness to understand others,
we have no reason to accept the hypothesis that we are faced with a belief system or language
that is incommensurable with or untranslatable into our own. It would be strange, given Rorty’s
anti-essentialism about language (not to mention rejection of the a priori/ a posteriori
distinction), that he would commit himself to the stronger, transcendental reading of Davidson.
Indeed, this attitude recommended by the pragmatic reading just mentioned is awfully close to
what Rorty calls ‘hermeneutics’ in PMN: ‘Hermeneutics sees the relations between various
discourses as those of strands in a possible conversation, a conversation which presupposes no
43
disciplinary matrix which unites the speakers, but where the hope of agreement is never lost so
long as the conversation lasts’ (318). 31
In ‘Cosmopolitanism without Emancipation: A Response to Jean-François Lyotard,’
Rorty brings Davidson’s attack on the notion of incommensurable conceptual schemes to
undermine such an appeal, though one made by Lyotard. While sympathetic to Lyotard’s
suspicion of the use of historical metanarratives in order to justify (or discount) social
institutions, Rorty is skeptical of what he takes to be Lyotard’s desire to abandon both liberal
parliamentary democracy and first-order stories of progress about such societies. For Rorty, even
after abandoning sweeping metanarratives, we can still praise Western liberal democratic
institutions ‘on the basis of invidious comparisons with suggested concrete alternatives’ (ORT,
211), and tell stories of moral and intellectual progress in order to ‘lift our spirits through utopian
fantasy, rather than to gird our loins with metaphysical weapons’ (212). Such stories will, as we
have seen, have to be ethnocentric in what Rorty takes to be an unproblematic sense. The moral
and intellectual progress made by the development and proliferation of liberal democratic
institutions is, for Rorty, tantamount to their tendency to produce more tolerant citizens, who rely
increasingly on ‘persuasion rather than force’ in their encounters with other cultures (214).32
Rorty reads Lyotard as throwing doubt on this pragmatic defense of liberal democracy on
the basis of a philosophical view about language. In particular the possibility of
incommensurable (i.e., untranslatable) discourses is supposed to call into question the very
31
I am indebted to Bruce Hunter for drawing my attention to the tension created by attributing a transcendental
reading of Davidson to Rorty, and for suggesting what a more pragmatic reading might look like. 32
‘Against [the] assimilation of the pragmatist’s inevitable ethnocentrism to Nazism,’ Rorty insists ‘that there is an
important difference between saying “We admit that we cannot justify our beliefs or our actions to all human beings
as they are at present, but we hope to create a community of free human beings who will freely share many of our
beliefs and hopes,” and saying, with the Nazis, “We have no concern for legitimizing ourselves in the eyes of
others.” There is a difference between the Nazi who says “We are good because we are the particular group we are”
and the reformist liberal who says “We are good because, by persuasion rather than force, we shall eventually
convince everybody else that we are.” (ORT, 214).
44
distinction between ‘persuasion’ and ‘force’ on which Rorty bases his defense. If members of
different cultures (or historical epochs) wield vastly diverse and incommensurable conceptual
apparatuses, then any attempt on the part of one culture to persuade another to adopt its values or
institutions will inevitably constitute a form of imperialist force (214). Moreover, if one takes
seriously the notion that one’s current conceptual scheme (or that of the culture of which one is a
member) may be radically different from that of one’s past and future (or past and future
cultures), then there is no sense to be made of intellectual and moral progress. This is precisely
the line of reasoning that Rorty attributes to Foucault in 1979. And while he is willing to employ
all the Davidsonian considerations mentioned above in order to undermine Lyotard’s critique of
the ‘pragmatist attempt to see the history of humanity as the history of the gradual replacement
of force by persuasion, the gradual spread of certain virtues typical of the democratic West’
(216), by the mid-1980’s, he is reluctant to make such criticisms of Foucault.
This raises several questions. Did Foucault actually subscribe to some form of conceptual
relativism? And if so, what does this amount to? That is, between what does he think it makes
sense to talk about incommensurability (i.e., conceptual schemes, vocabularies, paradigms)? And
finally, should it be established that Foucault does endorse some form of the incommensurability
thesis, does it simply dissolve into nonsense in light of Davidsonian considerations?
2.3 Incommensurability in Foucault
It is not difficult to find passages in Foucault’s writings that suggest an affirmative
answer to the first question raised above (i.e., whether or not Foucault actually endorsed some
form of conceptual relativism). One need not look any further than the first paragraph of the
preface to The Order of Things. The book, Foucault tells us, ‘first arose out of a passage in
45
Borges, out of the laughter that shattered, as I read the passage, all the familiar landmarks of my
thought—our thought that bears the stamp of our age and our geography’ (xv). The passage in
question references a taxonomy of animals taken from ‘a certain Chinese encyclopedia’ which
proposes that animals be classified as follows:
(a) belonging to the Emperor, (b) embalmed, (c) tame, (d) sucking pigs, (e) sirens, (f)
fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j)
innumerable, (k) drawn with a fine camelhair brush, (l) et cetera, (m) having just broken the
water pitcher, (n) that from a long way off look like flies (xv).
Foucault contends that in this taxonomy ‘the thing we apprehend is one great leap, the thing that,
by means of the fable, is demonstrated as the exotic charm of another system of thought, is the
limitation of our own, the stark impossibility of thinking that’ (xv). What does this impossibility
amount to? It is not, Foucault tells us, the sheer ‘oddity of the unusual juxtapositions’ mentioned
in Borges, but consists in ‘the fact that the common ground on which such meetings are possible
has been destroyed. What is impossible is not the propinquity of the things listed, but the very
site on which their propinquity would be possible’ (xvi). It is not a far stretch to interpret the
impossibility to which Foucault gestures, as a sort of incommensurability—a distance so great
between our own system of thought and the one to which this exotic taxonomy is at home. What,
exactly, does this amount to saying?
Consideration of the strange Borgesian taxonomy compels Foucault to wonder:
When we establish a considered classification, when we say that a cat and a dog resemble
each other less than two greyhounds do, even if both are tame or embalmed, even if both
are frenzied, even if both have just broken the water pitcher, what is the ground on which
we are able to establish the validity of this classification with complete certainty? On
what ‘table’, according to what grid of identities, similitudes, analogies, have we become
accustomed to sort out so many different and similar things (xix)?
46
The Order of Things attempts to answer these questions, not as Kant did by locating the
transcendental conditions of possible experience in the human mind, but by describing the
historical conditions through which different systems of knowledge or epistemes could emerge.
In a highly abstract sense, Foucault tells us that an episteme is an ‘epistemological field’ (xxii)
which ‘[i]n any given culture and at any given moment… defines the conditions of possibility of
all knowledge, whether expressed in a theory or silently invested in a practice’ (168). Following
Gary Gutting, we can say more precisely that for Foucault, an episteme involves not only the
conception of knowledge of a given period, but also a fundamental understanding of order (the
‘fundamental way in which it sees things connected to one another’ (Gutting, 139)), as well as its
particular conception of signs and language (140).
As a preliminary answer to the second question raised at the end of the previous section
(i.e., what, for Foucault, are the relata of incommensurability claims), one might construe
Foucault as committed to the view that there exists a sort of incommensurability between
epistemes. This gains plausibility when one considers that Foucault’s analyses explicitly attempt
to demonstrate that throughout history, epistemes undergo sharp ruptures; as they do so, entirely
different forms of knowledge emerge, eclipsing their predecessors. Whereas traditional historians
of knowledge, science, or ideas often describe the development of a particular empirical science
in terms of a smooth progression, or steadily increasing rationality, Foucault finds radical
discontinuities. The flip-side of this is that quite often, discourses, or disciplines usually thought
to be relatively disparate (e.g., economics, and linguistics), are for Foucault deeply connected at
the level of the episteme on which they arise. This is not to say that people are conscious of the
historical preconditions on which their knowledge depends. What makes The Order of Things so
revealing and yet so challenging is that it attempts to articulate that which, for any given
47
historical epoch, is presupposed in any body of knowledge but seldom (if ever) articulated.
Archaeology is, therefore, not concerned with what people know, or with evaluating the
epistemological status of the knowledge in a given period, but with, for example, what types of
statements were even candidates for knowledge, what types of reasoning were accepted as
capable of producing knowledge, or the unacknowledged presuppositions on which classification
depends.
Of course, describing at this level of abstraction Foucault’s aim in The Order of Things is
not very useful, and does very little to give sense to the claim that he holds that different
epistemes are incommensurable. It is his detailed historical descriptions that do all the work. In
this book, Foucault is concerned with elucidating three different epistemes—constituting what he
calls the Renaissance, Classical, and Modern periods. One of his central aims is to show that the
ways in which human beings came to know and reflect on language, the natural world, and
exchange were constrained in radically different ways throughout history. While an exposition of
The Order of Things in its entirety would allow me to spell out in great detail what exactly it
means to say that, for Foucault, the discontinuities between epistemes give rise to conceptual
incommensurability, it would probably provide more detail than is needed to establish this point.
Therefore, in what follows I will only summarize his account of the shift from the Renaissance to
the Classical episteme.
During the Renaissance, Foucault contends that, at least in the Western world, things
were ordered according to the notion of resemblance. He writes that,
[u]p to the end of the sixteenth century, resemblance played a constructive role in the
knowledge of Western culture. It was resemblance that largely guided exegesis and
interpretation of texts; it was resemblance that organized the play of symbols, made
48
possible knowledge of things visible and invisible, and controlled the art of representing
them (17).
Although he thinks that there were many conceptual resources through which the notion of
resemblance operated, Foucault identifies four particular notions that he thinks were most
essential to the Renaissance episteme: convenientia (convenience), aemulatio (emulation),
analogy, and the play between sympathy-antipathy. The first can be thought of as ‘a resemblance
connected with space in the form of a graduated scale of proximity’ (18). The spatial proximity
of things in the world was, during the Renaissance, not thought of as a mere external
relationship, but rather ‘the sign of a relationship, obscure though it may be’ (18). Thus,
according to Foucault, moss on the outside of shells, plants on the antlers of stags, or ‘a sort of
grass on the faces of men’ were all indications of a hidden relationship that human knowers
could come to recognize (18). In contrast to convenience, emulation is a form of resemblance,
that operates not though spatial proximity but at a distance. For example, ‘[t]here is something in
emulation of the reflection of a mirror… The human face, from afar, emulates the sky, and just
as man’s intellect is an imperfect reflection of God’s wisdom, so his two eyes, with their limited
brightness, are a reflection of the vast illumination spread across the sky by sun and moon’ (19).
The third form of resemblance, analogy, differs from the first two insofar as the phenomena it
deals with are not ‘the visible, substantial ones between things themselves; they need only be the
more subtle resemblances of relations’ (21). Foucault thinks that relationships of analogy
proliferate endlessly, and include, for example, ‘the relation of the stars to the sky… between
plants and the earth, between living beings and the globe they inhabit, between minerals such as
diamonds and the rocks in which they are buried, between sense organs and the face they
animate, between skin moles and the body of which they are the secret marks’ (21). Finally,
sympathy and antipathy are principles of mobility (23). The former explains the visible
49
movement of things (e.g., roots towards water, heavy objects towards the earth, sunflowers
toward the path of the sun), but also ‘gives rise to a hidden interior movement’ whereby a thing
involved in a relationship of sympathy will undergo qualitative change (e.g., ‘fire, because it is
warm and light, rises up into the air, towards which its flames untiringly strive; but in doing so it
loses its dryness’ (23).
This semantic web of resemblances that, for Foucault, manifests the particular way of
ordering things in the Renaissance, does not of itself make up the conception of knowledge given
by the episteme. A condition for their being knowledge was the existence of a mark or sign that
could make people aware of these forms of resemblance—what Foucault calls signatures. Indeed
‘knowledge of similitudes [i.e., resemblances] is founded upon the unearthing and decipherment
of signatures’ (26). It would, for instance, be impossible to recognize the ability of walnuts to
prevent ‘internal head ailments’ were it not for the mark of analogy—the resemblance between
the nut and the human brain—which indicates that there is a relation of sympathy (27). This
insight calls forth one of the essential features of the Renaissance episteme. Signatures (signs, or
marks) are themselves resemblances, and therefor make up part of the world. As Gutting puts it,
The sign of a sympathy may be an analogy, the sign of an analogy an emulation, of an
emulation, a convenience, of a convenience a sympathy once again. In this (and similar)
ways, Renaissance thought pursued knowledge of its world through an unending spiral of
linked resemblances, each a sign of the other. The system of the world and the system of
knowledge of the world had, accordingly, the same essential structure, that of a complex
of interconnected resemblances (Gutting, 142).
As is characteristic of much of The Order of Things, Foucault’s account of the movement
from the Renaissance episteme to that of the Classical period is both provocative and dramatic.
There was not, in the early part of the 17th century, simply a transformation in methods for
investigating the empirical world brought about by the emergence of new disciplines and areas of
50
inquiry, nor the inception of a new ‘curiosity’ in people’s minds, but rather ‘an essential rupture
in the Western world’ (50), whereby ‘the entire episteme of Western culture found its
fundamental arrangements modified’ (54). It would be a mistake, Foucault thinks, to be satisfied
with any historical account of this rupture framed in terms of the invention or discovery of the
scientific method along with its dispelling of superstition and magic from Western knowledge.33
Rather, he thinks that ‘what we must grasp and attempt to reconstitute are the modifications that
affected knowledge itself, at the archaic level which makes possible both knowledge itself and
the mode of being of what is to be known’ (54). Of the modifications that Foucault outlines,
there are two that appear to be the most salient. The first is a transformation of the way in which
things are ordered in the world, and the second is a reconfiguration of the relationship between
signs and the world.
Regarding the principle of ordering characteristic of the Classical episteme, Foucault
writes that ‘what has become important is no longer resemblances but identities and differences’
(50). Gutting puts this point as follows:
Resemblances are no longer regarded as expressing the true order of reality, an order that
is rather to be found in the structure of the elements into which things and their
resemblances can be analyzed. These elements are related not by vague and ambiguous
resemblances but by strict identities and differences (presence or absence of properties).
On the basis of these identities and differences, elements can be arranged in a series (e.g.,
from the simplest to the most complex) in terms of precise criteria (Gutting, 146).
This fundamental reconfiguration of the way in which the world is ordered, gave rise to a new
conception of knowledge described as the ‘project of a general science of order; a theory of signs
33
Foucault has a penchant for alluding to and opposing what might be thought of as ‘received views’ without
actually pointing to anyone who explicitly holds these views. Regarding the break between the epistemological
fields of the 16th
and 17th
centuries, one might imagine someone holding the view that Bacon’s Novum Organum, or
Descartes’ Rules for the direction of the mind, were discoveries that explain the emergence of new scientific
enterprises along with their associated bodies of knowledge. Foucault’s problem with such an historical account is
that it leaves unexamined what he identifies as a fundamental reconfiguration in the very conditions of possibility
for knowledge that had to take place in order for these thinkers to come to think about the world in the terms that
they did.
51
analyzing representations; the arrangement of identities and differences into ordered tables’ (OT,
71-2). Whereas the hallmark of renaissance knowledge can be thought of as the (potentially
endless) pursuit of recognizing resemblances, for the classical age, it is the construction of
general tables by analyzing mental representations on the basis of identities and differences.
While recognizing resemblances is essentially a process of composition—bringing things
together on the basis of a kind of similitude, establishing a ‘general science of order’ is
fundamentally a process of analysis—separating things by recognizing their different properties.
Moreover, as Gutting notes, this shift to the classical episteme brought about
changes in the character of the knowledge the mind attains. Because it dealt with an
unending chain of resemblances, Renaissance knowledge was necessarily incomplete and
merely probable. By contrast, the elements revealed by Classical analysis could be
completely enumerated and exhaustively understood. Accordingly, Classical knowledge
could, at least in principle, attain complete certitude (147).
The notion of ‘mental representations’ is the key to understanding the second
modification. Foucault writes that ‘[o]n the threshold of the Classical age, the sign ceases to be a
form of the world; and it ceases to be bound to what it marks by the solid and secret bonds of
resemblance and affinity’ (58). Before the seventeenth century, knowledge of the world involved
the recognition of signs (i.e., signatures) that formed part of systems of resemblance. Crucially,
and as we have seen, these signs were themselves resemblances, and thus were conceived as a
part of the world. This is what makes their bonds solid. They were secret, because though their
raison d’être was to be discovered, ‘they did not need to be known in order to exist: even if they
remained silent, even if no one were to perceive them, they were just as much there’ (59). In the
Classical era, by contrast, signs were detached from the world and became co-extensive with
thought (65). Moreover, signs (i.e., ideas, perceptions, sensations) were no longer construed in
52
terms of resemblance (i.e., signatures understood as signs resembling other signs), but in terms of
their ability to directly represent their object. As Foucault puts it:
signs are now set free from that teeming world throughout which the Renaissance had
distributed them. They are lodged henceforth within the confines of representation, in the
interstices of ideas, in that narrow space in which they interact with themselves in a
perpetual state of decomposition and recomposition (67).
This alteration of the way in which signs were understood had several important consequences.
In contrast to the Renaissance, because they came to be understood as mental representations,
signs were no longer ‘secret’ in the sense that their reality was assured independent of their being
discovered by a knower. On the classical conception the ‘sign does not wait in silence for the
coming of a man capable of recognizing it: it can be constituted only by an act of knowing’ (OT,
59). For Foucault the ‘most fundamental’ feature of classical signs has to do with the relation to
what they signify. On the renaissance conception, the answer to the question of what links a sign
to what it signifies is obvious: another sign. Yet when signs become, for the classical age, cut off
from the material world—‘lodged within the confines of representation’—how are they related to
their content? And what guarantees that this relationship is accurate? Foucault suggests that the
classical response to the first question renders the second moot. A striking feature of classical
representation was their ‘duplicated’ function. To be a sign, it is not enough for an idea,
perception, or mental image to merely signify something (i.e., another idea, perception or mental
image), but it must also show its own signifying power in the act of representing. As Foucault
puts it, a representation ‘can become a sign only on condition that it manifests, in addition the
relation that links it to what it signifies. It must represent; but that representation, in turn, must
also be represented within it’ (64). One surprising conclusion that Foucault draws from the
classical conception of representation is that it ‘precludes even the possibility of a theory of
signification’(65). This is because ‘all representations are interconnected as signs; all together,
53
they form, as it were, an immense network; each one posits itself in its transparency as the sign
of what it represents; and yet… no specific activity of consciousness can ever constitute a
signification’ (65-6). As Gutting puts it, ‘representation cannot be represented’ (153). But if this
is the case, it could never occupy a position in an ordered system of representations, and be
understood ‘in terms of its relationship of identity and difference to the other elements (Gutting,
153).
Together, these features ‘constituted an area of empiricity in the Classical age that had
not existed until the end of the Renaissance, and that was destined to disappear early in the
nineteenth century’ (OT, 72). To make his case, Foucault points to three disciplines—general
grammar, natural history, and the analysis of wealth—and tries to show how each of these
empirical sciences arose from, and were constrained in similar ways by the Classical episteme.
That is, it was on the basis of a particular conception of signs as mental representations, and of
knowledge as tables ordered on the basis of the identities and differences of properties that these
disciplines could emerge. Here, I will only touch briefly on Foucault’s contrast between natural
history and Renaissance writings on nature, as they are especially illuminating. He writes that,
Until the time of Aldrovandi, History was the inextricable and completely unitary fabric
of all that was visible of things and of the signs that had been discovered or lodged in
them: to write the history of a plant or animal was as much a matter of describing its
elements or organs as of describing the resemblances that could be found in it, the virtues
that it was thought to possess, the legends and stories with which it had been involved, its
place in heraldry, the medicaments that were concocted from its substance, the foods it
provided, what the ancients recorded of it, and what travelers might have said of it. The
history of a living being was that being itself, within the whole semantic network that
connected it to the world (129).
What for us, seem like obvious distinctions between the physical descriptions of an animal, and
for instance, the stories in which it happens to be mentioned, would have for the Renaissance
been seen as arbitrary or perhaps nonsensical. When struck with how bizarre it seems that
54
Aldrovandi would include in his account of serpents their form and description alongside gods to
which they are dedicated, the riddles in which they are involved, or their significance in
dreams34
, we are tempted to think that he was somehow being unreasonable. It was not,
according to Foucault, that half a century later when natural historians began making the
distinction between physical descriptions of organisms and their place in mythology, that they
did so on the basis of deeper and more far-reaching knowledge about the world, or because
Reason had been awakened within them. What made possible such a distinction was that signs
were no longer ‘part of things themselves… [but] modes of representation’ (129).
With this sample of Foucault’s archaeological project in The Order of Things at hand, we
are in a better position to evaluate the questions that concluded the previous section. It seems
that, at least in his archaeological period, Foucault did in fact subscribe to some form of
conceptual relativism. In a general sense, we can say he held that different epistemes were
incommensurable with one another. More precisely, the very ways things were ordered in the
world, or the way in which signs and language were conceived, gave rise to forms of knowledge
that were in very precise ways, completely disparate. Given the Rennaissance conception of
resemblance, Foucault has identified a concrete sense in which it was impossible for intellectual
projects such as general grammar, or natural history to emerge.35
These disciplines could, for
Foucault, only arise given a radical rupture in the episteme which constituted, among other
34
Foucault presents us with a list of headings from the chapter ‘On the Serpent in General’ form Aldrovandi’s
Historia serpentum et draconum which includes : ‘equivocation (which means the various meanings of the word
serpent), synonyms and etymologies, differences, form and description, anatomy, nature and habits, temperament,
coitus and generation, voice, movements, places, diet, physiognomy, antipathy, sympathy, modes of capture, death
and wounds caused by the serpent, modes and signs of poisoning, remedies, epithets, denominations, prodigies and
presages, monsters, mythology, gods to which it is dedicated, fables, allegories and mysteries, hieroglyphics,
emblems and symbols, proverbs, coinage, miracles, riddles, devices, heraldic signs, historical facts, dreams,
simulacra and statutes, use in human diet, use in medicine, miscellaneous use’ (OT, 39). 35
‘Generally speaking’ he asks, rhetorically, ‘what does it mean, no longer being able to think a certain thought? Or
to introduce a new thought?’ (OT, 50).
55
things, a conception of knowledge that involved a particular way of ordering objects, and a
particular paradigm of signification.
Foucault’s commitment to the conceptual relativity of epistemes serves at once to confirm
Rorty’s initial characterization of his work, and at the same time, to indicate what appears to be a
substantive philosophical disagreement between them. While this outcome is of itself important
insofar as it appears to have been overlooked by commentators, not to mention by Rorty himself
(despite his early essay on Foucault), it is only the springboard to the major claim I hope to
advance. In the next chapter I shall argue that disagreement over the possibility of
incommensurability can explain the apparent political disagreement between the two men. On
the one hand, I will argue that Rorty’s reformist liberalism finds philosophical support, from his
rejection of conceptual relativism. On the other hand, I will argue that Foucault’s apparent
refusal to admit that liberal societies constitute instances of progress, as well as his unwillingness
to sketch utopias, is supported by his openness to the possibility of incommensurability between
our current socio-political institutions and practices, and those of the past and future.
56
Chapter Three: Politics of Incommensurability
In chapter one, I outlined what I took to be the central features of Rorty’s political
disagreement with Foucault. I claimed that this disagreement amounted to a double refusal: on
the one hand, a backward-oriented refusal to call present liberal social institutions and practices
progress, and on the other, a forward-oriented refusal to offer concrete proposals for ameliorating
or replacing such institutions and practices. I showed that, at least on Rorty’s construal, this
radical stance towards politics was, for Foucault, not tethered to any philosophical position, but
rather, resulted from what can broadly be understood as ‘external’ factors.
Without entirely discounting the ‘external’ factors discussed by Malecki, the aim of this
chapter is to show that Rorty and Foucault’s political divergence can be accounted for on the
basis of their disagreement over conceptual relativism. My strategy will be to examine separately
both sides of what I have been calling Foucault’s double-refusal. In section 3.2 I shall argue that,
while Foucault’s unwillingness to call liberal societies progress stems from his openness to the
imminent possibility of incommensurable discourses, precisely the contrary is true for Rorty.
That is, Rorty’s pragmatic defense of liberalism is only plausible if liberal societies can be placed
within first-order narratives which present such societies as having made progress with respect to
the past; yet the viability of such narratives is purchased by his acceptance of Davidson’s
arguments against conceptual relativism. In section 3.3 I examine two consequences of
Foucault’s conceptual relativism and argue that they support his unwillingness to sketch utopias.
3.1 Liberal Progress: Incommensurable Views
Given that political questions in general, and questions about liberalism in particular do
not seem to be at all raised in Foucault’s archaeological writings, one might wonder: What does
refusing to praise liberal institutions have to do with his commitment to conceptual relativism?
57
After all, as an archaeologist, he appears to be interested primarily in understanding the historical
conditions of possibility of the human sciences. How could his identification of ruptures and
discontinuities in the epistemes underwriting such disciplines possibly explain his refusal to
commend liberal institutions? My argument that it does shall involve two steps. First I show that
Foucault thought it was misguided to ask whether or not the shift from one episteme to another
counted as progress. Second, I argue that since Foucault’s archaeological project extends beyond
investigating epistemes (e.g., to political discursive practices), he had similar reasons to eschew
the notion of progress regarding political systems. A central part of this archaeological expansion
is that Foucault’s later notion of ‘governmentality’ (also called political rationalities) functions in
the same way as his earlier notion of epistemes, but involves political discursive practices rather
than the human sciences.
In many ways, the bodies of knowledge made possible by the Renaissance episteme
were, as Foucault put—‘plethoric yet absolutely poverty-stricken’ (30). This is because, ‘[b]y
positing resemblance as the link between signs and what they indicate… sixteenth-century
knowledge condemned itself to never knowing anything but the same thing, and to knowing that
thing only at the unattainable end of an endless journey’ (30). Given this deficiency, one might
think that with the rupture marking the beginning of the Classical period, human knowledge
progressed—by ordering the world according to the identities and differences of representations,
it became possible to have much more systematic accounts of nature than were possible when
resemblances reigned. Though tempting, Foucault is reluctant to draw such a conclusion. While
comparatively, the Classical period was able to shed some of the Renaissance limitations on
knowing, it did so only by setting up new deficiencies. Foucault makes this point by contrasting
58
the work of the 17th
-century natural historian John Jonston with that of Ulisse Aldrovandi
(mentioned above). He writes,
When Jonston wrote his Natural history of quadrupeds, did he know any more about
them than Aldrovandi did, a half-century earlier? Not a great deal more, the historians
assure us. But that is not the question. Or, if we must pose it in these terms, then we must
reply that Jonston knew a great deal less than Aldrovandi. The latter, in the case of each
animal he examined, offered the reader, and on the same level, a description of its
anatomy and of the methods of capturing it; its allegorical uses and mode of generation;
its habitat and legendary mansions; its food and the best ways of cooking its flesh.
Johnston subdivides his chapter on the horse under twelve headings: name, anatomical
parts, habitat, ages, generation, voice, movements, sympathy and antipathy, uses, medical
uses. None of this was omitted by Aldrovandi, and he gives us a great deal more besides.
The essential difference lies in what is missing in Jonston (OT, 129).
As we have seen, Foucault thinks that this difference was made possible, in part, by the
ontological separation of words and things—the shift from resemblance to representation. Yet
even though there is a sense in which Jonston’s natural history provides its reader with ‘less
knowledge’ than Aldrovandi’s, insofar as it simply omits certain things, Foucault clearly does
not want to engage in such a discussion. Besides the superficial quantitative comparison, their
incommensurability at the level of the episteme makes evaluative comparison futile. As Ian
Hacking puts it, ‘[w]e cannot reason as to whether alternative systems of reasoning are better or
worse than ours [or each other], because the propositions to which we reason get their sense only
from the method of reasoning employed. The propositions have no existence independent of the
ways of reasoning towards them’ (HO, 175).
This suggests that Foucault’s reluctance to offer a positive (or negative) evaluation of
changes in epistemes in general, results from his having viewed such changes as
incommensurable. In order to show that this line of reasoning can help explain his attitude
towards liberal institutions, I shall argue that although Foucault initially applied his
archaeological analyses to those discursive practices which aspire to scientificity, he believed
59
they could be applied more broadly to socio-political discursive practices. By elaborating on this
‘expanded’ scope of archaeology, I will suggest that the same sort of considerations which
explain Foucault’s reluctance to attribute progress to the human sciences can explain his
reluctance to attribute progress to liberal institutions (and thus, his disagreement with liberals
such as Rorty). Since he viewed both modern knowledge-producing practices (i.e., the human
sciences) and political discourse as incommensurable with those of the past, the question of
whether or not to evaluate them in terms of progress is simply beside the point.
Foucault first explicitly discusses the possibility of expanding the scope of archaeology to
socio-political domains in the final section of chapter six of The Archaeology of Knowledge.
‘Must archaeology,’ he asks, ‘be—exclusively—a certain way of questioning the history of the
sciences... or has it provided an outline, on the basis of a particular example, of forms of analysis
that may have a much wider application’ (192)? At the time, Foucault admits that he is ‘not
sufficiently advanced in his task to answer this question’ but suggests some promising avenues
for future research. After foreseeing a couple of possibilities—an archaeology of sexuality36
and
of painting—Foucault considers a third archaeological domain. He writes that
It seems to me that one might also carry out an analysis of the same type on political
knowledge. One would try to show whether the political behavior of a society, a group, or
a class is not shot through with a particular, describable discursive practice. This
positivity would obviously not coincide either with the political theories of the period or
with economic determinations: it would define the element in politics that can become an
object of enunciation, the forms that this enunciation may take, the concepts that are
employed in it, and the strategic choices that are made in it (194).
36
Regarding an archaeology of sexuality, Foucault writes: ‘Such an archaeology would show, if it succeeded in its
task, how the prohibitions, exclusions, limitations, values, freedoms, and transgressions of sexuality, all its
manifestations, verbal or otherwise, are linked to a particular discursive practice. It would reveal, not of course as
the ultimate truth of sexuality, but as one of the dimensions in accordance with which one can describe it, a certain
‘way of speaking’; and one would show how this way of speaking is invested not in scientific discourses, but in a
system of prohibitions and values. An analysis that would be carried out not in the direction of the episteme, but in
that of what we might call the ethical (193)’. Less than a decade later, Foucault would, of course, write the first
volume of The History of Sexuality. At some point in the course of these years he would decide to approach the
project genealogically, thus apparently abandoning this earlier vision.
60
Such a project, if successful, would circumvent some of the problems which Foucault implicitly
ascribes to a Marxist approach to political theory (which were prevalent in 1960’s France).
Namely, it would avoid the need to ‘pass through the authority of an individual or collective
consciousness in order to grasp the place of articulation of a political practice and theory’
(194).37
A ‘political’ archaeology would simply ‘analyse the formation and transformations of a
body of knowledge’ (194), and attempt to ‘explain the formation of a discursive practice and a
body of revolutionary knowledge that are expressed in behavior and strategies, which give rise to
a theory of society, and which operate the interference and mutual transformation of that
behavior and those strategies’ (195).
While Foucault’s brief and tentative remarks only scratch the surface of what an
archaeology of political knowledge would amount to, his confidence in the plausibility of such a
project is suggestive. Although he would go on to do a genealogy of sexuality, Foucault would
eventually follow through with his intention to do an archaeology of political knowledge. As
Colin Koopman rightly points out, Foucault’s later work on political rationality was ‘explicitly
archaeological in orientation’ (2008, 342). In his 1977-8 and 1978-9 lectures at the College de
France Foucault developed something very close to an archaeology of political knowledge. But
unlike his analysis of epistemes undertaken in The Order of Things, his lectures investigate what
he called ‘governmentality.’
In one sense, governmentality means ‘art of government’ or ‘governmental rationality’. It
is a term which is intended to capture a type of discourse in which governmental practices are
taken up, reflected upon, and altered—while at the same time, producing and limiting political
subjects and institutions. As governmental practices become represented in discourse and
37
These criticisms, of course, chime with Foucault’s problematization of the history of ideas that presupposes a
constituting human subject.
61
reflected upon, the very domain or field deemed legitimate for governmental consideration,
along with the criteria by which successful or unsuccessful governance is determined, form a
type of rationality. This rationality in turn enables and limits the ways in which different types of
political subjects and institutions can be produced and acted upon by governmental practices
(which themselves become the objects of representation and reflection).
Just as epistemes are historically situated and undergo ruptures, so too, different
governmental rationalities emerge at definite points in history. For example, Foucault contends
that throughout the Middle Ages, the aim of the sovereign qua governor was to ensure the
salvation of his or her subjects in the afterlife. Thus, sovereign legitimacy was founded in a
certain ethical or religious relationship to the divine. Foucault contends that in the 16th
century, a
new form of governmental rationality emerged, characterized by reflection on the state as a
‘relatively autonomous’ reality (BB, 4). Rather than concerning itself with the salvation of its
subject’s souls, the state became organized around what Foucault calls the principles of raison
d’état. Not only did the goals of government shift toward enrichment via monetary
accumulation, strengthening through increased population, and engagement in permanent
competition with other states, but these came to be rationalized as legitimizing principles for the
state (5). For Foucault, the organization of this new art of government around the principles of
raison d’état gave rise to important limitations on the objectives of governance. Insofar as
individual states were forced to recognize the aspirations and interests of other competing states,
they encountered external self-limitations in terms of their own objectives which found
expression in foreign policy. Most notably, and in contrast to the Middle Ages, it was no longer
acceptable for any state to have as its aim to be the ‘total and global empire at the end of time’
(6). Various political theories of ‘European balance’, and the emergence of military-diplomatic
62
treaties, were manifestations of the self-limiting principle of states, in recognition of their
perpetual competition with one another. In contrast to this (external) self-limitation of states
Foucault thinks that according to the principles of raison d’état there was no such limitation with
respect to a state’s internal policy, or what he calls the police state. That is, ‘there is no limit to
the objectives of government when it is a question of managing a public power that has to
regulate the behavior of subjects’ (7). While the police state, or internal state policy of raison
d’état may have encountered various types of juridical resistance throughout the sixteenth and
seventeenth centuries, it was not until the middle of the 18th
century that there emerged an
internal regulation of governmental rationality (10). For Foucault, the appearance of this internal
regulation marks an important transformation from governmental rationality in accordance with
the principles of raison d’état to modern liberal governmental rationality in accordance with the
principles of political economy. He summarizes the transformation as follows:
From the sixteenth and seventeenth centuries, and even before until the middle of the
eighteenth century, there was a whole set of practices and tax levies, customs charges,
manufacturing regulations, regulations of grain prices, the protection and codification of
market practices, and so on… [A]ll of these practices were certainly reflected on, but on
the basis of different events and principles of rationalization. From the middle of the
eighteenth century it becomes possible to establish a reasoned, reflected coherence
between these different practices… a coherence established by intelligible mechanisms
which link together these different practices and their effects, and which consequently
allows one to judge all these practices as good or bad, not in terms of a law or moral
principle, but in terms of propositions subject to the division between true and false (BB,
18).
Through the principles of political economy, it became possible for government to reflect upon
and limit its modes of intervention with respect to an increasingly connected web of political and
economic practices, not simply in terms of legitimacy (i.e., juridical or moral valances), but
according to a new ‘regime of truth’. To illustrate this point, Foucault contrasts how the market
was conceived in the Middle Ages up through the seventeenth century, with how it was
63
conceived along the lines of the type of political rationality characteristic of modern liberalism.
In the former case, the market was ‘essentially a site of justice’ (30). That is, it was ‘invested
with extremely prolific and strict regulations’ that functioned to ensure a just distribution of
goods, as well as to protect against theft and fraud. Moreover, the price of goods was
conceptualized both theoretically and practically, in terms of justice—a price ‘was to have a
certain relationship with work performed, with the needs of the merchants, and, of course, with
the consumers’ needs and possibilities’ (30). By the middle of the 18th
century, there was a
transformation in governmental rationality, whereby the market was no longer viewed as a site of
justice, but rather as a site of truth. As Foucault puts it,
On one hand, the market appeared as something that obeyed and had to obey “natural,”
that is to say, spontaneous mechanisms. Even if it is not possible to grasp these
mechanisms in their complexity, their spontaneity is such that attempts to modify them
will only impair and distort them. On the other hand—and this is the second sense in
which the market becomes a site of truth—not only does it allow natural mechanisms to
appear, but when you allow these natural mechanisms to function, they permit the
formation of a certain price that Boisguilbert will call the “natural” price, the physiocrats
will call the “good price,” and that will later be called the “normal price,” that is to say, a
certain price…which will adequately express the relationship, a definite, adequate
relationship between the cost of production and the extent of demand (31).
In terms of the political rationality characteristic of the Middle Ages (up through the 17th
century), it was possible to ask whether or not a price was just—but to ask whether or not a price
was ‘true’ would have been regarded as nonsensical or wildly metaphorical. Conversely, given
the political rationality that emerged in the middle of the 18th
century, it became possible to
question whether or not a price was ‘true’ (i.e., according to its ability to reflect the value of a
product determined by the relationship between supply and demand); yet this shift also meant
that considerations of justice were no longer relevant to the price of a particular good. These
conceptual differences indicate that what counts as a plausible and effective strategy for
64
evaluating, or criticizing governmental decision making is in a sense, dependent on the
conceptual possibilities made available by a particular governmentality. In the Middle Ages, it
might have made sense to criticize a sovereign for failing to ensure that the price of grain was
low enough that the poorest peasants could afford to feed themselves (i.e., failure to ensure that
the market functioned as a site of distributive justice); yet in the late 18th
century, such a
criticism would, according to Foucault’s analyses, appear misguided (in fact, an attempt made by
the sovereign to alter the price of grain would itself be grounds for criticism).38
With this expanded conception of archaeology in view, it is now possible to see how
considerations analogous to those which would motivate his rejection of narratives of progress in
the natural sciences apply to Foucault’s conception of liberalism. Taken as dependent on a
particular political rationality, Foucault would come to view liberalism as in many concrete ways
incommensurable with those of the past. Had Foucault been asked explicitly whether or not he
thought liberal institutions and the governmental rationality on which they depend were better
than those of the 16th
or 17th
centuries, I suspect that he would have reacted in the same way he
did to the question of whether natural historians ‘knew more’ than their 16th
-century
‘predecessors.’ There is simply no space for the notion of ‘progress’ in Foucault’s political
thought, at least insofar as it is informed by his archaeological analyses.
By contrast, because Rorty’s commitment to liberalism is (to paraphrase Barry Allen)
dependent on how it is described, and especially, as it is described in comparison with past,
present and future societies, he very much needs to preserve conceptual space for a notion of
progress. Yet, what might be thought of as the traditional understanding of political progress is
38
Another example of this sort of difference is that, in the Middle Ages, committing a mortal sin could have counted
as grounds for criticizing a sovereign if it prevented them from ensuring the salvation of their subjects. Many
contemporary politicians sin on daily basis (e.g., commit perjury, get divorced, espouse atheism), yet this is not
(usually) seen as relevant to whether or not they govern effectively.
65
unavailable to him. Rather than saying, for instance, that liberal democracies are superior to
fascist dictatorships because they allow for the true realization of human nature, or that ‘liberal
freedom has a “moral privilege”’ which is absent in the value system of fascist dictatorships,
Rorty regards ‘the justification of liberal society simply as a matter of historical comparison with
other attempts at social organization—those of past and those envisaged by utopians’ (53). This
consideration lies at the heart of Rorty’s insistence that ‘liberal culture needs an improved self-
description rather than a set of foundations’ (CIS, 52).
In Contingency, Irony and Solidarity, Rorty builds his notion of progress (i.e., political,
moral, scientific) on the idea that different societies employ different vocabularies. A given
political vocabulary will supply members of a society, among other things, with the descriptive
and normative resources for advancing and debating political proposals. Vocabularies are, of
course, open-ended, and constantly evolving. Human beings are always coming up with new
ways of speaking—for Rorty, inventing metaphors— sometimes as attempts to cope with social
change (i.e., institutional, technological), sometimes as attempts to affect social change, and
sometimes without any specifiable reason at all. The taking-up of these new ways of speaking—
that is, the literalisation of metaphors—forms the basis of new vocabularies. Thus, for Rorty,
‘progress, for the community as for the individual, is a matter of using new words as well as of
arguing from premises phrased in old words’ (48-9). There is, of course, no neutral meta-
vocabulary that would supply anyone with absolute criteria for judging or comparing
vocabularies, but we can always make retrospective judgments and comparisons on the basis of
the criteria supplied by our present vocabulary. Rorty makes this point by likening vocabularies
to tools. Human beings value both for their ability to help in solving practical problems. Just as
we might choose one tool over another because it better serves our present needs, he suggests
66
that we should evaluate the gradual replacement of one vocabulary by another in the same way.39
Yet, unlike tools, whose usefulness can almost always be specified before their creation,
by contrast, the creation of a new form of cultural life, a vocabulary, will have its utility
explained only retrospectively. We cannot see Christianity or Newtonianism or the
Romantic movement or political liberalism as a tool while we are still in the course of
figuring out how to use it. For there is no clearly formulatable ends to which it is a
means. But once we figure out how to use the vocabularies of these new movements, we
can tell a story of progress, showing how the literalization of certain metaphors served
the purpose of making possible all the good things that have recently happened. Further,
we can now view all these good things as particular instances of some more general good,
the overall end which the movement served. This latter process was Hegel’s definition of
philosophy: “holding your time in thought.” I construe this to mean “finding a description
of all the things characteristic of your time of which you most approve, with which you
unflinchingly identify, a description which will serve as a description of the end toward
which the historical developments which led up to your time were means (55).
Given the sketch of Rorty’s postmodern bourgeois liberalism presented in chapter one, the
aspects of liberal societies of which Rorty most approves should not come as a surprise. They
include the widespread desire to reduce suffering and humiliation, the belief that disputes should
be resolved through persuasion rather than force, and the hope of creating a classless and
casteless society. Perhaps more importantly, as we have seen he thinks that ‘contemporary liberal
society already contains the institutions for its own improvement’ (63). Because of this, he
suspects that
Western social and political thought may have had the last conceptual revolution it needs.
J.S. Mill’s suggestion that governments devote themselves to optimizing the balance
between leaving people’s private lives alone and preventing suffering seems to me pretty
much the last word. Discoveries about who is being made to suffer can be left to the
workings of a free press, free universities, and enlightened public opinion—enlightened,
for example by books like Madness and Civilization and Discipline and Punish, as well
as those like Germinal, Black Boy, The Road to Wigan Pier, and 1984 (63-4).
39
In chapter one of CIS Rorty’s aim is to substitute the vocabularies-as-tools metaphor, for the vocabularies-as-
jigsaw-puzzles metaphor. According to the latter, advances in political, moral, and scientific vocabularies are
explained by their having come to
67
This is, of course, not to deny the need for political revolutions. Indeed, Rorty is the first to
admit that ‘it is hard to imagine a diminution of cruelty in countries like South Africa, Paraguay,
and Albania without a violent revolution’ (63, fn. 21).
Rorty’s strategy for developing a notion of progress, if successful, would allow him to
praise liberalism while excising from its vocabulary those remnants of Enlightenment
rationalism intended to supply philosophical foundations for liberal societies. Yet his account
raises some important questions. Given that his figure of the “ironist” is someone who realizes
‘that anything can be made to look good or bad by being redescribed’ (73), what is Rorty to
make of someone who comes along and redescribes the very description with which he so
unflinchingly identifies? This is, of course, the position that Rorty attributes to Foucault—the
ironist who is unwilling to be a liberal. On the picture I advanced in chapter one, part of this
disagreement is over the question of whether or not, compared to past social forms, the benefits
of liberal democracies outweigh their harmful effects. Moreover, Rorty chooses to interpret this
disagreement as stemming from external factors, rather than on the basis of an underlying
philosophical tension.
Although he does not state it explicitly, I want to suggest that Rorty’s account of progress
(as just outlined) tacitly makes use of Davidson’s argument against incommensurability. As we
have seen, Rorty is able to avoid witless relativism and ‘justify’ his commitment to liberal
democracies, solely on the basis of concrete comparisons with other social forms. Against
someone who does not share his optimism—someone who would choose to describe such
societies in darker, more insidious terms—Rorty can only reiterate what he takes to be the virtues
and aims of liberalism and insist that we have yet to come up with anything better. But what if
someone were to object that such comparisons were either in principle, or for particular reasons,
68
impossible? For example, one might contend that the vocabulary of liberal politics is, in some
sense, incommensurable with the political vocabularies employed by other societies (past,
present, and future). If such a contention could be made to look plausible, it would prove
disastrous for Rorty’s pragmatic defense of liberalism because it would preclude the very
possibility of meaningful comparison on which his claims to superiority are based. He would not
be justified in telling retrospective stories of progress—they would be too ‘Whiggish’ to be of
any use. As we have seen, Rorty’s agreement with Davidson allows him to circumvent this
possible objection to his view—he is confident that we will always be able to meaningfully
compare liberal societies with their predecessors, because he is convinced that the very idea of
incommensurable vocabularies is a non-starter.
3.2 On the Very Idea of Sketching Utopias
The above considerations suggest not only that Foucault had philosophical reasons for
refraining from praising 20th
-century liberal societies as instances of progress, but that Rorty’s
pragmatic defense of liberalism gains its plausibility from a tacit dismissal of conceptual
relativism. What about the second dimension of their political disagreement—that is, Foucault’s
refusal to sketch utopias? In this section I shall discuss two consequences of Foucault’s
conceptual relativism, and argue that when understood in a political sense, they create problems
for those who offer concrete political proposals. The first consequence, given the sort of
conceptual relativism to which Foucault subscribes, is that it is always possible that what is
intended to be a radically new proposal (i.e., scientific theory, political utopia) will simply
occupy another space within the current system of possibilities (i.e., episteme, political
rationality). The second consequence is that because ruptures at the archaeological level are not
consciously available to the agents who experience them, it is always possible that an historical
69
rupture could take (or have just taken) place, rendering obsolete any given political proposal
which was dependent on the previous system of possibilities. I shall argue that, when taken
seriously, the first consequence poses problems for revolutionary political proposals, and the
second for reformist ones such as those which Rorty advocates.
Before spelling out these implications of Foucault’s conceptual relativism, it is worth
briefly examining some of the passages where Foucault states his forward-looking refusal. The
first of these sets of remarks appears in a 1971 interview, where Foucault discusses (among other
things) his involvement in the Group d’information sur les prisons (GIP). It is in this interview
that Rorty finds his ‘least favorite thing’ that Foucault ever said in print; thus it is the site on
which the political divergence is most apparent between the two thinkers. One of Foucault’s
interlocutors poses the ‘tiresome question… [which] must be faced eventually: what replaces the
system?’ (230). To which Foucault replies:
I think that to imagine another system is to extend our participation in the present system.
This is perhaps what happened in the history of the Soviet Union: apparently, new
institutions were in fact based on elements taken from an earlier system—the Red Army
reconstituted on the model of the Czarist army, the return to realism in art, and the
emphasis on the traditional family morality (230-1).
This remark is not merely a momentary aberration, but a position Foucault would maintain on
more than one occasion.40
That same year, he would make effectively the same point during a
highly publicized debate with the American linguist and political theorist Noam Chomsky.
For Chomsky, there are two intellectual tasks when it comes to politics. The first is the
‘somewhat more abstract and philosophical task’ (42) of coming up with a socio-political theory
on the basis of insights gleaned of human nature. In particular, Chomsky’s research into
linguistics led him to conclude that since ‘a fundamental element of human nature is the need for
40
For example, recall the interview mentioned in chapter one, wherein according to Rorty, Foucault attempts to
‘outradical’ some Maoists, by making essentially the same point about the establishment of people’s courts.
70
creative work…without the arbitrary limiting effect of coercive institutions’ (37), we should
outline and devote our political energy towards establishing a social form which best realizes this
essence.41
The second task is to come to understand ‘the nature of power and oppression and
terror and destruction in our own society’ (42).
Foucault applauds the second ‘intellectual task’—indeed, for him it is more ‘immediate’
and ‘urgent’ than anything else (40)—while expressing deep disquiet about the first. He not only
admits ‘to not being able to define, nor for even stronger reasons to propose, an ideal social
model’ (40), but finds the attempt to ground such a proposal in an appeal to human nature deeply
problematic. ‘But isn’t there a danger here? ’ he asks:
If you say that a certain human nature exists, that this human nature has not been given in
actual society the rights and the possibilities which allow it to realize itself… doesn’t one
risk defining this human nature—which is at the same time ideal and real, and has been
hidden and repressed until now—in terms borrowed from our society, from our
civilization, from our culture? (43)
As in the interview cited above, Foucault appeals to how early 20th
-century socialism, in
attempting to liberating human nature which had been alienated under capitalism, ended up
simply ‘transposing’ bourgeois values into Soviet society. Believing that they had imagined a
new system, the Soviet visionaries had merely extended their participation in the present.
Ultimately, Foucault concludes, ‘it is difficult to say exactly what human nature is. Isn’t there a
risk that one will be led into error? Mao Tse-Tung spoke of bourgeois human nature and
proletarian human nature, and he considers that they are not the same thing’ (43-4).
Despite Rorty’s repugnance towards these sets of remarks, it is worth noting that he
would have to side with Foucault rather than Chomsky over the question of whether we should
41
Indeed, Chomsky himself thought that anarcho-syndicalism (i.e., ‘a federated, decentralized system of free
associations, incorporating economic as well as other social institutions’ (38)) was just the sort of social arrangement
we humans need to flourish.
71
try to base our utopic sketches on an account of human nature. Perhaps the reason he regards
these remarks as so ‘fatuous’ is because he interprets Foucault as attempting to opt out of liberal
politics in favour of some sort of radical quietism—where all we can do is attempt to burn to the
ground our present institutions along with their associated conceptual frameworks; all the while
doing everything we can to make sure that the institutions of tomorrow resemble as little as
possible the ashes of yesterday’s. I would suggest, however, that it is more plausible to interpret
Foucault’s claim that ‘to imagine another system is to extend our participation in the present
system’ as a cautionary reminder to someone insistent on completely revolutionizing their
current social system: that it is always possible that the radical changes one seeks to bring about
will merely transpose current social problems into a different key.
This isn’t to say that Foucault is denying the possibility of political revolution—in fact as
we have seen, he is as attuned as anyone to the history of radical transformations in the ways
human beings have thought about themselves and their relation to the world. Rather, it is to say
that Foucault is throwing doubt on the idea that a revolution can be thought up in advance,
especially using the concepts of the present. We can see how this latter interpretation of
Foucault’s refusal to sketch a utopia dovetails with his commitment to conceptual relativism by
considering how his archaeological analyses reveal that theories, or even entire disciplines that
are typically held to be in opposition (and thus incompatible), are actually a part of the same
system of possibilities.42
As Ian Hacking puts it,
the systems of thought to which Foucault addresses himself are not constituted by a
unified set of beliefs advanced by a person or a school, Indeed, he has a teasing device
42
One might object that the examples in question do not support the notion that Foucault’s refusal to sketch utopias
stems from his commitment to conceptual relativism—after all, what Foucault is essentially claiming is that often,
those who take themselves to have thought up a revolutionary new set of practices and institutions (e.g., the Soviets
creating the Red Army) engage in self-deception insofar as they fail to see how their innovation borrows elements
from the prevailing practices and institutions. While these examples involve something supposedly
incommensurable turning out to be commensurable, my point is that this very line of reasoning depends on
Foucault’s acceptance of the notion of incommensurable frameworks in the first place.
72
that I call Foucault’s fork, which surprises us by stating that competing bodies of belief
have the same underlying rules of formation. Once there was a memorable contrast
between the taxonomic System of Linnaeus and the Method of Adanson. We now have
little difficulty in supposing that these antagonistic enterprises are part of the same web of
possibilities and alternatives, but some of us are more startled to read that positivism and
phenomenology are equally constituted by a common underlying organization (90).
Given that we have already seen how Foucault’s archaeological method can be expanded to the
political domain through the notion of political rationalities, it is not difficult to see the
similarities between the insights motivating his ‘fork’ and those motivating his refusal to sketch
revolutionary political proposals. For example, consider the (supposed) opposition between the
‘method’ and ‘system’ approach to natural history that Foucault describes in The Order of
Things.43
As a natural historian operating within the Classical episteme, Buffon’s ‘method’ was
not a radically novel alternative to Linnaeus’s ‘system’, but was rather an extension of his
participation in the available system of conceptual and theoretical possibilities. On my account,
Foucault is making an analogous point, for example, when he suggests that Chomsky’s attempt
to imagine an alternative to advanced capitalist societies risks an unwitting continuation of the
very elements he seeks to do away with. His claim is that Chomsky’s anarcho-syndicalism would
43
The first, and perhaps (by today’s standards) more familiar approach known as the System operates by selecting ‘a
finite and relatively limited group of characteristics, whose variations and constants may be studied in any individual
entity that presents itself’ (OT, 139). Proponents of the system would begin by isolating such a group of
characteristics called a structure, on the basis of which to establish each individual species (e.g., of plant or animal)
on a taxonomic table. Foucault notes that the structure was both arbitrary and relative. Arbitrary in the sense that it
‘deliberately ignores all differences and all identities not related’ to it (140); and relative to the number of variables
used to analyse a given structure. The second approach, called the method, began with as exhaustive a description of
a plant or animal as possible ‘leaving out none of its parts and determining all the values that the variables have
derived from it’ (142). Then further species would be described just as exhaustively, ‘but with the one difference
that nothing that has been mentioned in the first description should be repeated’ (142). This process would be
repeated, and ‘by arranging the later and progressively more sparse descriptions around the earlier ones… we shall
be able to perceive, through the original chaos, the emergence of the general table of relations’ (142). Obviously,
these two approaches to natural history bear striking differences. As Foucault notes, while it is only possible for
there to be a single method, any number of different ‘systems’ could be invented—according to the degree of
precision etc. Moreover, natural historians took themselves to be in disagreement about the correct or best way of
classifying the natural world. Regardless, Foucault maintains that ‘both system and method rest upon the same
epistemological base… [whereby] knowledge of empirical individuals can be acquired only from the continuous,
ordered, and universal tabulation of all possible differences’ (144).
73
merely be another space within the same underlying system of political possibilities. Indeed,
toward the end of their debate Foucault states that,
Contrary to what you [Chomsky] think, you can’t prevent me from believing that these
notions of human nature, of justice, of the realization of the essence of human beings, are
all notions and concepts which have been formed within our civilization, within our type
of knowledge and our form of philosophy, and that as a result form part of our class
system; and one can’t, however regrettable it may be, put forward these notions to
describe or justify a fight which should—and shall in principle—overthrow the very
fundamentals of our society. This is an extrapolation for which I can’t find the historical
justification. That’s my point… (57-8)
One advantage of taking seriously this political consequence of Foucault’s conceptual
relativism, is that it allows his refusals to sketch political utopias to be understood as part of a
more general pattern that emerges throughout much of his post-1970 writings—what might be
termed his cautioning of the politically engaged. That is, rather than construing the remarks
mentioned at the outset of this section as invoking some sort of radical quietism (as Rorty seems
to do), they can, and should be understood as cautioning his interlocutors about the dangers of
putting forth a revolutionary political project (i.e., of sketching utopias). Perhaps the most
widely-known example of this cautioning is exemplified in his polemic against the ‘repressive
hypothesis’ in volume I of The History of Sexuality. Just like the French Maoists and Noam
Chomsky, proponents of the repressive hypothesis (i.e., French Freudo-Marxists of the 1960’s)
are attempting to sketch a socio-political utopia; and while arguably well-intentioned, they all
run an analogous risk. The latter subscribe to a narrative, according to which, through a series of
prohibitions, silencings, censorships, and denials, sexuality has been repressed in bourgeois
societies since the 17th century. Consequently, sexual (and in a sense, political) liberation from
this repression is thought to require an ‘irruption of speech’ about sex; that by speaking so
74
openly and often about sexuality, one will subvert and transgress this repression, thereby
realizing new sexual and political freedoms.
Foucault is critical of this model for several reasons.44
Part of his point—and this is what
I take to be analogous to his critique of Chomsky—is that by presupposing the model of
repression supplied by the repressive hypothesis, sexual liberation movements run the risk of
sketching political utopias that are not only consonant with, but leave them unable to discern, the
very mechanisms of power they seek to challenge. For example, advocating for increased open
public discussion about sex, or perhaps that individuals seek the nature of their own sexual
identity through psychotherapy, are not only unpromising avenues for achieving sexual or
political liberation (if there even is such a thing), but can be seen as part of the same ‘historical
framework’ within which Western societies have attempted to speak the truth about sex through
the proliferation of sexual discourses.
By taking seriously this first implication of Foucault’s conceptual relativism (i.e., that
what is taken to be a radically new proposal may simply turn out to be another possibility within
the same system) we can see that pace Rorty, there is more to Foucault’s claim that to imagine
another system is to extend our participation in the present than adherence to a contingent
fashion. Not only can his remarks be understood as part of his broader cautioning of the
politically engaged, but they stem from Foucault’s archaeological insights worked out in The
Order of Things.
44
Foucault raises ‘three serious doubts’ concerning this repressive hypothesis. First, he is skeptical of the historical
fact that there has indeed been a repression of sexuality for the last few centuries, especially in the form of
censorship, silencing, and prohibition. Rather, his historical investigations suggest that the last three centuries have
given rise to ‘a veritable discursive explosion’ regarding sexuality. Second, Foucault doubts that, in general, the
model of repression as prohibition, censorship and denial is apt for capturing the ways in which power is exercised,
especially in contemporary Western societies. Finally, and for our purposes most importantly, he suggests that the
common strategic response by proponents of the repressive hypothesis is misguided. Foucault asks, ‘Did the critical
discourse that addresses itself to repression come to act as a roadblock to a power mechanism that had operated
unchallenged up to that point, or is it not in fact part of the same historical network as the thing it denounces (and
doubtless misrepresents) by calling it “repression”? (10).
75
There is, however, a second consequence of Foucault’s conceptual relativism that poses
similar problems for those interested in sketching utopias. And unlike the consequence just
discussed which most obviously creates challenges for revolutionary political proposals (such as
Chomsky’s, or the Freudo-Marxist proponents of the repressive hypothesis), it would seem to
pose worries for reformist projects of the sort Rorty would endorse.
Given the high degree of precision with which Foucault is able to identify discontinuities
between epistemes, one might wonder: What happens to those unfortunate enough to be caught
in the middle of a rupture? One might think that Foucault owes us some sort of psychological or
phenomenological account of what it is like to start one’s career, for instance, as a natural
historian, and end up as a biologist. As we have seen, Foucault steadfastly refuses to do this—
insisting on offering descriptions of the positive unconscious of an epoch. This is one of the
crucial points at which Foucault’s archaeological project differs from Kuhn’s account of
paradigm change. For Kuhn, the psychological experience of researchers who undergo changes
in scientific paradigm is likened to a gestalt-shift. He writes that
[a]t times of revolution, when the normal-scientific tradition changes, the scientist’s
perception of his environment must be re-educated—in some familiar situations he must
learn to see a new gestalt. After he has done so the world of his research will seem, here
and there, incommensurable with the one he had inhabited before. That is… [a] reason
why schools guided by different paradigms are always slightly at cross-purposes (Kuhn,
112).
Moreover, whereas Kuhn maintains that during a scientific revolution, researches must decide
whether or not to abandon the old paradigm in favour of a new one, Foucault’s position seems to
76
be that scientists are not presented with a conscious choice as to whether or not to take up an
emerging episteme. The movement from one episteme to another is as silent as it is profound.45
If this is correct, does it not follow that the threat of a rupture is always and everywhere,
persistent? Not only does Foucault’s notion of the positive unconscious seem to entail that we
could find ourselves in the position analogous to Aldrovandi’s—where the system of possibilities
within which we come to know ourselves and the world is on the verge of collapse; but it
suggests, perhaps with equal probability, that such a collapse may have just taken place, and that
the separation between our current selves and (what we take to be) our intellectual predecessors
is radically severed in ways that only future archaeologists shall be in a position to notice. When
transposed into a political key, this second implication of Foucault’s conceptual relativism
invites a sort of skepticism towards the sort of reformist political utopias that Rorty thinks
liberals ought to sketch. Put in terms of Foucault’s notion of ‘political rationalities,’ the worry is
that if the possibility of a rupture at the level of political rationality is always imminent, and such
ruptures are not consciously discernable by political agents, then there is no way of determining
whether or not the political vocabulary we find ourselves to be employing is an obsolete relic of
a past. For Foucault, it is possible for a discursive formation in which a particular political utopia
is sketched to undergo a radical break. Not only does this entail the possibility that the utopia is
framed in terms that are incommensurable with the new discursive formation, but the agents who
proposed it would not be in a position to discern that a transformation has taken place.
There are, of course, many different ways in which this incommensurability might arise.
For example, at the end of the Middle Ages, a viable political utopia might have involved global
domination. On Foucault’s analysis, since the political rationality of the time had not yet set
45
This is presumably why, in his discussion of the archive in The Archaeology of Knowledge, Foucault suggests that
there must be a certain degree of historical distance in order for the archaeologist to even discern such
discontinuities.
77
external limitations on a state’s ambitions, this would have been a plausible political project.
Suppose, however, that a rupture had just taken place, and that a new political rationality had
come to eclipse the one within which the original political utopia of global domination had been
formulated. Because—following Foucault’s analyses—this new political rationality placed
external constraints on what was considered to be appropriate aims of governing, global
domination would no longer be regarded as an acceptable political proposal. It would be
analogous to insisting, during the Classical period, that the legends in which a given animal is
mentioned, or what its presence in dreams symbolizes, be included in its natural historic
taxonomy. If, however, like epistemes, transformations in political rationalities are not
(immediately) noticeable by agents involved, then the futility of holding to such a utopia would
not even be recognizable by its proponents. Those who encounter ruptures at the level of political
rationality run the risk of not even being able to determine whether or not the political utopias
they sketch are even relevant.
In the previous section I argued that by ruling out in advance the possibility of conceptual
relativism, Rorty was able to appeal to an account of progress on which his pragmatic defense of
liberalism was based. Likewise, I contend that the same consideration (i.e., against conceptual
relativism) allows him to dismiss the sort of skepticism about the continuing viability of liberal
reformism. Both Rorty and Foucault agree that there are socio-historical limitations on what can
be meaningfully expressed. Insofar as they cannot be stepped outside of, Foucault’s notion of
discursive practices and Rorty’s vocabularies amount to pretty much the same thing. To sketch a
political utopia is to do so within a vocabulary; to imagine a socio-political system is to be
involved in a discursive practice. Moreover, vocabularies resemble discursive practices in that
both supply a normative framework within which to assign ends for our utopic sketches, to
78
determine the criteria of their success or failure, as well as what counts as an appropriate means
to achieve them. And while both vocabularies and discursive practices are historically
contingent, whereas Rorty thinks it makes no sense to speak of incommensurable vocabularies,
Foucault accepts examples of incommensurable discursive practices.46
It is precisely because of
this difference that Rorty is able to circumvent the skepticism towards sketching political
utopias, which as I have argued, follows from Foucault’s conceptual relativism. For Foucault, it
is always possible that the vocabulary in which a particular political utopia is sketched, could be
eclipsed by another, incommensurable political vocabulary. This is simply not possible on
Rorty’s account.
In this chapter I have taken up both sides of the political disagreement between Foucault
and Rorty, and attempted to show that it is grounded in their divergence over the question of
conceptual relativism. First, I argued that while Foucault’s refusal to call liberal societies
progress can be understood as an extension of the fact that the archaeological insights he
developed in The Order of Things left no room for an account of progress between epistemes
(owing to their incommensurability), Rorty’s liberalism gains its plausibility from his rejection
of the possibility of incommensurable vocabularies. Second, I examined two consequences of
Foucault’s commitment to conceptual relativism, and argued that they provide support for his
reluctance towards sketching concrete political proposals.
46
Although, up to now I have been holding to the claim that, for Foucault, epistemes are incommensurable, since he
seems to think that discursive practices depend on an episteme, it follows that when the latter undergo ruptures, so
do the former.
79
Conclusion
In the final section of Chapter Three I argued that, pace Rorty, Foucault’s reluctance to
sketch utopias can also be understood as stemming from two consequences of his conceptual
relativism. The first was skepticism towards the sort of political projects that purport to bring
about social forms that are held to be radically different than those of the present. The second
was skepticism about the ability of present political tools—that is, our current political
vocabularies—to be applicable to future cases. What motivated this skepticism, I claimed, was
the fact that Foucault seems to allow for the possibility of political rationalities—what I suggest
can be read as the political analogue to epistemes—to undergo historical ruptures, and that these
ruptures are not immediately discernable to those involved. We might imagine someone who,
oblivious to such a ‘rupture,’ is ‘stuck’ employing the prior system of possibilities. Unlike the
Chomskyan or the Freudo-Marxist, who seeks radical change, this person sketches reformist
utopias, but by making use of a system of possibilities which has become defunct. By failing to
recognize the incommensurability between their political proposals and the new political
rationality in which they are not at home, the reformist’s project is condemned at best to have
marginal effects, at worst to be completely irrelevant.
As a way of winding down my discussion, I would like to offer a gloss on what I take to
be a practical upshot of this last (admittedly abstract) point. That is, rather than push the question
of who got the problem of conceptual relativism right, I want to conclude by showing how this
latter consequence of Foucault’s commitment to conceptual relativism—which I have claimed
lends support to his refusal to sketch reformist political utopias—led him to approach the
development of neoliberalism in the 20th
century in a manner that is markedly different than that
of Rorty.
80
In the previous chapter I offered a brief account of Foucault’s work on political
rationality that was developed in great detail in his course lectures from 1977-9, but I did not
mention what is very likely the most significant upshot of these lectures. Despite its title, The
Birth of Biopolitics is not really about biopolitics at all, but about ‘governmentality.’ More
precisely, it is predominantly about the historical development of a particular type of
governmentality: that of neoliberalism. By now, Foucault’s strategy of tracing the historical
development of governmental rationalities has engendered an entire field of study,
governmentality studies. A likely reason for this fecundity is that the approach was initially so
successful. In what has to be one of the most remarkable instances of intellectual prescience,
Foucault set his sights on analyzing the nascent economic doctrines that had been developing in
post-war Germany and the United States—the so-called Ordo liberal and Chicago schools of
economics, respectively. Years before Ronald Reagan’s and Margaret Thatcher’s names would
become symbolic for a campaign of policies that would dismantle North American and European
social-safety nets, completely overhaul monetary policy, and put into motion a series of free-
trade mechanisms that would ultimately revolutionize the face of the global economy, Foucault
was, for whatever reason, drawn to the fringe group of economic thinkers (Milton Friedman,
Friedrich Hayek, and Gary Becker) whose theories would provide the basis for these policies.
Whether or not Foucault foresaw the scale to which this movement would eventually proliferate
in the decades after his death, or the deleterious social effects it would bring in its wake, his
insights suggest an attunement to which Rorty was arguably oblivious.
Part of Foucault’s attunement is reflected in his realization that the emergence of
neoliberal political rationality indicated a transformation in—if not a rupture from—the political
rationality characteristic of liberalism. That is, the shift from traditional liberalism to
81
neoliberalism (which Foucault correctly perceived to be taking place in the mid-20th
century)
involved a transformation in how the nature, limits, and proper function of governmental practice
were to be conceived.47
This involved an incredibly far reaching transformation in the domain,
methodology and concepts of economics, but more importantly, set limitations on acceptable
policy proposals and political interventions.48
As one might expect, given the account offered in
the past three chapters, Foucault refrains from offering an explicit evaluation of the development
of neoliberalism, not to mention any specific political recommendations about how to respond to
it. Rather, by investigating the historical development of neoliberal political rationality—
especially its break from that of liberalism, Foucault can be thought of as having alerted others to
47
Jason Read offers a helpful summary of Foucault’s distinction between liberal and neoliberal governmentality. He
writes, ‘[c]lassical liberalism focused on exchange, on what Adam Smith called mankind’s tendency to “barter,
truck, and exchange.” It naturalized the market as a system with its own rationality, its own interest, and its own
specific efficiency, arguing ultimately for its superior efficiency as a distributor of goods and services. The market
became a space of autonomy that had to be carved out of the state through the unconditional right of private
property… [Whereas] [c]lassical liberalism makes exchange the general matrix of society… [n]eoliberalism,
according to Foucault, extends the process of making economic activity a general matrix of social and political
relations, but takes as its focus not exchange but competition (27)’. The shift from exchange to competition as the
characteristic feature of neoliberal governmentality has at least two important implications. First, whereas for
liberalism the market was conceived of as a site of exchange governed by the natural mechanisms of supply and
demand which needed to be protected against governmental intervention, under neoliberalism the market became
understood as an artificial site of competition which required continuous state intervention through regulatory and
organizational actions. Although these tactics of intervention are irreducible to specific economic and political
policies shared by the German, French and American forms of neoliberalism, a common feature of neoliberal
governmentality involves the extension of market rationality increasingly to every facet of social life. Drawing on
the work of Chicago school economists (most notably Gary Becker), Foucault points out how the vocabulary and
rationality of micro economics (i.e., decision theory) began to permeate areas of social and political life which had
never before been conceived of as economic domains. What were traditionally understood as moral or social
issues—from criminality and alcoholism, to family relations and education—became the object of economic
analysis in mid-twentieth century America. Second, this shift implied a transformation in the type of economic
subject—homo economicus—treated and produced by neoliberal governmentality. For many political economists of
the eighteenth century, homo economicus was ‘the person who must be left alone’ (BB, 270) to pursue her own,
inalienable interests. However by the mid-twentieth century, the economic subject became understood as ‘eminently
governable’ by virtue of being someone who simply ‘responds systematically to modifications in the variables of the
environment’ (270). As Foucault puts it, ‘[f]rom being the intangible partner of laissez-faire, homo economicus now
becomes the correlate of a governmentality which will act on the environment and systematically modify its
variables’ (270-1). 48
This realization motivated yet another example of what I called Foucault’s ‘cautioning of the politically engaged’.
In the Birth of Biopolitics, he warned of the so-called ‘great state phobia’—the view that responsible political action
requires one to combat, perhaps at all costs, the ever-expanding encroachment of state policies on individual liberty.
Foucault cautioned that leftists inflicted with such a phobia were merely ‘following the direction of the wind and
that in fact, for years and years, an effective reduction of the state has been on the way’ (BP, 191).
82
the waning possibility of sketching the sort of utopias that may have been plausible given the
liberal political rationality that was dominant in the early decades of the 20th
century.
Rorty, I would suggest, found himself confronting the same historical emergence of
neoliberalism, but responded to it in a very different way than Foucault. Throughout the 1980’s
until his death, Rorty witnessed the harmful effects that neoliberal policies were having on
American society. Although he was able to foresee some of the long-term consequences of these
policies, Rorty’s repeated calls for a return to reformist New Deal-style leftism were virtually
ineffective. In contrast with the downbeat but optimistic tone of his early political writings, one
is struck within his later work by the anxiety and pessimism with which he regarded the future of
liberal democratic societies. In a 2003 interview, his interlocutor remarks that,
In most of your earlier political writing, you talked about the need for hope and patriotic
pride—faith in the idea that our country’s wrongs might be redeemed—as precursors to
genuinely progressive political strides. This Old Leftist-style pride was contrasted to the
sullen defeatism of the academic left—people like Foucault, who once said that to
imagine an alternative future is just to further our participation in the total corruption of
the present system. More recently, however, you’ve quoted Gore Vidal and the poet
Robinson Jeffers, who wrote that America was “heavily thickening to empire.” Have you
become less sanguine about our political future?
To which Rorty replies:
Well, at this point you have to quote Gramsci: “pessimism of the intellect, optimism of
the will.” If I had to lay bets, my bet would be that everything is going to go to hell, but,
you know, what else have we got except hope?
I have always found it surprising that Rorty never uses the word ‘neoliberalism’ in print:
It is as though he was at once fully aware of and entirely oblivious to the phenomenon. On a few
occasions he uses the term ‘neo-conservativism’, but mostly, it seems, to describe the emergence
of evangelical Christianity in the mainstream of the Republican Party. For all the concerns he
had about the effects of increased deregulation and globalization of the labour markets, and
83
despite his disquiet about tax cuts for the wealthy and erosion of the social safety-net, Rorty
seems to have been unable (or unwilling) to see these phenomena as the effects of a changing
hegemonic body of economic and political doctrines.
I take this example to illustrate the different attitudes with which Rorty and Foucault
responded to (more or less) the same political situation. Rorty’s insistence on the viability of
those political policies that (in his eyes) had been largely successful in post-war America, stands
in contrast to Foucault’s apparent unwillingness to prescribe any concrete political project. As I
have been suggesting more generally, I think that this difference in attitudes can be understood as
stemming from their underlying disagreement regarding conceptual relativism. In Foucauldian
terms, one might say that Rorty failed to recognize that his reformist liberalism could find no
place within neoliberal governmentality: that the vocabulary within which his concrete utopias
were stated was incommensurable with the emerging political vocabulary that defined the nature
and limits of governmental practice in the final two decades of the 20th
century (and persists to
this very day). Although I suspect that Rorty would have admitted that his proposals for
returning to welfare-state politics were largely ineffective, I doubt that he would have seen any
point in the Foucauldian claim that neoliberal political rationality is incommensurable with
liberal political rationality. While there is room on his account for the idea that a particular
political vocabulary could be the wrong tool for the job, as we have seen, he doesn’t leave room
for the idea that two vocabularies can be incommensurable with one another. By contrast, given
that Foucault does seem to have been committed to such an idea, it is understandable that he
would have been skeptical of the sort of early post-war liberalism to which Rorty subscribed. He
viewed it as part of a system of political possibilities that was well on its way to becoming
eclipsed.
84
Over the course of the preceding three chapters I have presented a novel approach to the
political disagreement between Richard Rorty and Michel Foucault. On this account, for Rorty,
Foucault’s unwillingness to be a liberal amounts to a ‘double refusal,’ first, to regard liberal
institutions as instances of historical progress, and second, to offer concrete political proposals
(or, as Rorty puts it, to sketch utopias). Moreover, in Chapter One I suggested that Rorty would
often frame this political disagreement in terms of external (i.e., social and biographical) factors,
rather than in terms of underlying philosophical differences. Against this background, I
developed a proposal in which the political disagreement between the two men could be
understood in terms of their diverging views with respect to the thesis of conceptual relativism.
In the second chapter, I contrasted Rorty’s rejection of the notion of incommensurable
conceptual schemes with Foucault’s commitment—espoused in The Order of Things—to the
existence of historically situated systems of thought (epistemes) that undergo periodic ruptures. I
claimed that for Foucault, when epistemes undergo such ruptures the bodies of knowledge to
which they give rise are incommensurable with one another. In the third chapter I suggested
several ways in which this philosophical disagreement over conceptual relativism could explain
the political disagreement outlined in chapter one. To show that Foucault’s refusal to admit of
liberal political progress stems from his commitment to conceptual relativism, I offered what can
be thought of as an argument by analogy: in The Order of Things, Foucault suggests that
characterizing shifts from one episteme to the next as instances of epistemic progress is
misguided; and since he would eventually extend his archaeological analysis to political
discourse (in his writings on political rationalities), we can infer that the same considerations
would apply to his views about liberalism. By contrast, I argued that Rorty’s pragmatic defense
of liberalism, in part, gains its plausibility by ruling out in advance the possibility of conceptual
85
relativism. Finally, I examined two implications of Foucault’s conceptual relativism I showed
how they both provide support for his reluctance to prescribe concrete political proposals.
86
Bibliography
Allen, Barry. “After Knowledge and Liberty: Foucault and the New Pragmatism.” Foucault’s
Legacy. Ed. Carlos G. Prado. London: Continuum, 2009.
Chomsky, Noam, and Michel Foucault. The Chomsky-Foucault Debate: On Human Nature. New
York: The New Press, 2006.
Davidson, Donald. “A Coherence Theory of Truth and Knowledge.” In Subjective,
Intersubjective, Objective: Philosophical Essays Volume 3. New York: Oxford
University Press, 2001.
--- “On the Very Idea of a Conceptual Scheme.” Proceedings and Addresses of the American
Philosophical Association 47 (1973-4): 5-20.
Dieleman, Susan. “Revisiting Rorty: Contributions to a Pragmatist Feminism.” In Hypatia: A
Journal of Feminist Philosophy. 25th Anniversary Special Issue 25.4 (2010): 891-908.
Dreyfus, Hubert L., and Paul Rabinow. Michel Foucault: Beyond Structuralism and
Hermeneutics. 2nd
ed. University of Chicago Press, 1983.
Festenstein, Matthew, and Simon Thompson eds. Richard Rorty: Critical Dialogues. Polity,
2001.
Foucault, Michel. Discipline and Punish: the Birth of the Prison. Trans. Alan Sheridan. New
York: Vintage Books, 1977.
--- Language, Counter-Memory, Practice: Selected Essays and Interviews. Ed. Donald F.
Bouchard. Trans. Donald F. Bouchard and Sherry Simon. Ithaca: Cornell University
Press, 1977.
--- Power/Knowledge. Ed. Colin Gordon. New York: Pantheon Books, 1980.
--- Security, Territory, Population: Lectures at the Collège de France 1977--1978. Ed. Michel
Senellart, trans. Graham Burchell. Palgrave Macmillan, 2007.
--- The Archaeology of Knowledge & the Discourse on Language. Trans. A. M. Sheridan Smith.
New York: Pantheon Books, 1972.
--- The Birth of Biopolitics: Lectures at the Collège de France (1978-79). Ed. Michel Senellart,
trans. Graham Burchell. Palgrave Macmillan, 2008.
--- The Foucault Reader. Ed. Paul Rabinow. New York: Pantheon, 1984.
--- The History of Sexuality Volume 1: An Introduction. Trans. Robert Hurley. Vintage Books,
1990.
--- The History of Sexuality Volume 2: The Use of Pleasure. Trans. Robert Hurley. New York:
Vintage Books, 1990.
87
--- The Order of Things: An Archaeology of the Human Sciences. New York: Vintage Books,
1994.
--- The Politics of Truth. Eds. Sylvère Lotringer, and Lysa Hochroth. New York: Semiotext (e),
1997.
Gross, Neil. Richard Rorty: The Making of an American Philosopher. Chicago: The University
of Chicago Press, 2008.
Gutting, Gary. Michel Foucault’s Archaeology of Scientific Reason. New York: Cambridge
University Press, 1989.
Fraser, Nancy. “Foucault: A “Young Conservative”?” Ethics 96.1 (Oct. 1985): 165-184.
Habermas, Jurgen, and Seyla Ben-Habib. “Modernity Versus Postmodernity.” New German
Critique 22 (Winter, 1981): 3-14.
Hacking, Ian. Historical Ontology. Cambridge: Harvard University Press, 2002.
Halperin, David, M. Saint Foucault: Towards a Gay Hagiography. New York: Oxford
University Press, 1995.
Harvey, David. A Brief History of Neoliberalism. New York: Oxford University Press, 2005.
Hoy, David Couzens. Foucault: A Critical Reader. New York: Basil Blackwell, 1986.
Koopman, Colin. “Foucault’s Historiographical Expansion: Adding Genealogy to Archaeology.”
Journal of the Philosophy of History 2 (2008): 338-62.
--- “Two Uses of Genealogy: Michel Foucault and Bernard Williams.” In Foucault’s Legacy. Ed.
Carlos G. Prado. London: Continuum, 2009.
Kuhn, Thomas S. The Structure of Scientific Revolutions. 2nd
ed. Chicago: The University of
Chicago Press, 1970.
Malecki, Wojciech. ““If Happiness Is Not The Aim Of Politics, Then What Is?”: Rorty Versus
Foucault.” Foucault Studies 11 (2011): 106-125.
McWhorter, Ladelle. Bodies and Pleasures: Foucault and the Politics of Sexual Normalization.
Indiana University Press, 1999
Miller, James. The Passion of Michel Foucault. New York: Doubleday, 1993.
Mouffe, Chantal., ed. Deconstruction and Pragmatism.New York: Routledge, 1996.
Nilsson, Jacob, and Sven-Olov Wallenstein, eds. Foucault, Biopolitics and Governmentality.
Sodertorn Philosophical Studies, 2013.
88
Oksala, Johanna. "Feminism and Neoliberal Governmentality." Foucault Studies 16 (2013): 32-
53.
Prado, C. G. “Foucault, Davidson, and Interpretation.” In Foucault and Philosophy. Ed. Timothy
O’Leary and Christopher Falzon. Wiley-Blackwell, 2010.
Ramberg, Bjørn T. "Strategies for Radical Rorty (‘… but is it progress?’)." Canadian Journal of
Philosophy 23.sup1 (1993): 223-246.
Read, Jason. "A Genealogy of Homo-economicus: Neoliberalism and the Production of
Subjectivity." Foucault Studies (2009): 25-36.
Rorty, Richard. Achieving Our Country: Leftist Thought in Twentieth-Century America.
Cambridge, MA.: Harvard University Press, 1998.
--- Consequences of Pragmatism (Essays: 1972-1980). Minneapolis: University of Minnesota
Press, 1982.
--- Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press, 1989.
--- Essays on Heidegger and Others: Philosophical Papers. Volume 2. Cambridge University
Press, 1991.
--- “Foucault and Epistemology.” In Foucault: a Critical Reader. Ed. David Couzens Hoy.
Oxford-New York: Basil Blackwell, 1986.
--- Objectivity, Relativism and Truth: Philosophical papers, Volume 1. Cambridge: Cambridge
University Press, 1991.
--- "Paroxysms and Politics." Salmagundi 97 (1993): 61-68.
--- Philosophy and the Mirror of Nature. Princeton: Princeton University Press, 1979.
--- Philosophy as Cultural Politics: Philosophical Papers, Volume 4. Cambridge: Cambridge
University Press, 2007.
--- “The Professor and the Prophet” Review of The American Evasion of Philosophy: A
Genealogy of Pragmatism. By Cornel West. Transition 52 (1991): 70-8.
--- Truth and Progress: Philosophical Papers, Volume 3. Cambridge: Cambridge University
Press, 1998.
--- "Response to Ernesto Laclau." In Deconstruction and Pragmatism (1996): 69-76.
Rorty, Richard, Derek Nystrom, and Kent Puckett. Against Bosses, Against Oligarchies: A
Conversation with Richard Rorty. Chicago: Prickly Paradigm Press, 2002.
89
Rorty, Richard, and Eduardo Mendieta. Take Care of Freedom and Truth Will Take Care of
Itself: Interviews with Richard Rorty. Stanford: Stanford University Press, 2006.